project management ournal · tel: +1-610-356-4600, ... project management institute global...

PR O J E C TMA N A G E M E N TJO U R N A L

Research Quarterly, Volume 38, Number 2 JU N E 2007

Making Project Management Indispensable for Business Results™

Published by

5ManagingKnowledge andLearning in ITProjects: A ConceptualFramework andGuidelinesfor Practice

Blaize Horner Reich

18Project ManagementProcessesAcross Borders: AComparisonof EU-US CorporateSubsidiaryProject Activities

Philippe Byosiere andDenise J. Luethge

30Soft SkillsQuantification(SSQ) for ProjectManagerCompetenciesEdward Muzio,Deborah J. Fisher, Erv R. Thomas, andValerie Peters

39ProjectScheduling:Improved Approach toIncorporateUncertaintyUsing BayesianNetworksVahid Khodakarami,Norman Fenton, andMartin Neil

50We’re Not In Kansas Anymore,Toto: Mapping theStrange Landscapeof ComplexityTheory, and ItsRelationship toProjectManagementTerry Cooke-Davies,Svetlana Cicmil, LynnCrawford, and Kurt Richardson

62Leadership Behaviors inMatrixEnvironments

Jerry Wellman, PMP

84ContainingTransaction Costsin ERPImplementationThrough Identificationof StrategicLearning ProjectsAnnika Andersson andRalf Müller

75Paradoxes of Control:The (Electronic)Monitoringand Reporting System of theDutch High SpeedAlliance (HSA)Kees Boersma, Sytze F. Kingma, andMarcel Veenswijk

93Project ManagementResearch—The Challenge andOpportunityAaron J. Shenhar andDov Dvir

PR O J E C TMA N A G E M E N TJO U R N A LThe Professional Research Journal of the Project Management Institute

Volume 38, Number 2

JUNE 20073 From the Editor

Christophe N. Bredillet, PhD, MBA, Ingénieur EC Lille

PAPERS

5 MANAGING KNOWLEDGE AND LEARNING IN IT PROJECTS: A CONCEPTUAL FRAMEWORK AND GUIDELINES FOR PRACTICE

Blaize Horner Reich

18 PROJECT MANAGEMENT PROCESSES ACROSS BORDERS: A COMPARISON OF EU-US CORPORATE SUBSIDIARY PROJECT ACTIVITIES

Philippe Byosiere and Denise J. Luethge

30 SOFT SKILLS QUANTIFICATION (SSQ) FOR PROJECT MANAGER COMPETENCIES

Edward Muzio, Deborah J. Fisher, Erv R. Thomas, and Valerie Peters

39 PROJECT SCHEDULING: IMPROVED APPROACH TO INCORPORATE UNCERTAINTYUSING BAYESIAN NETWORKS

Vahid Khodakarami, Norman Fenton, and Martin Neil

50 WE’RE NOT IN KANSAS ANYMORE, TOTO:MAPPING THE STRANGE LANDSCAPE OF COMPLEXITY THEORY, AND ITS RELATIONSHIP TO PROJECT MANAGEMENT

Terry Cooke-Davies, Svetlana Cicmil, Lynn Crawford, and Kurt Richardson

62 LEADERSHIP BEHAVIORS IN MATRIX ENVIRONMENTS

Jerry Wellman, PMP

75 PARADOXES OF CONTROL: THE (ELECTRONIC) MONITORING AND REPORTINGSYSTEM OF THE DUTCH HIGH SPEED ALLIANCE (HSA)Kees Boersma, Sytze F. Kingma and Marcel Veenswijk

84 CONTAINING TRANSACTION COSTS IN ERP IMPLEMENTATION THROUGHIDENTIFICATION OF STRATEGIC LEARNING PROJECTS

Annika Andersson and Ralf Müller

93 PROJECT MANAGEMENT RESEARCH—THE CHALLENGE AND OPPORTUNITY

Aaron J. Shenhar and Dov Dvir

100 Cover to Cover—Book Reviews

Kenneth H. Rose PMP

104 Calendar of Events

104 Index of Advertisers

EditorChristophe N. Bredillet, PhD, MBA,

Ingénieur EC Lille

PMI PUBLISHING STAFF

PublisherDonn Greenberg; [email protected]

Contributing EditorRoberta Storer; [email protected]

Project Management Journal AssociateNatasha Pollard; [email protected]

Publications PlannerBarbara Walsh; [email protected]

Bookstore [email protected]

General e-mail:[email protected]

Journal in print:[email protected]

Book Review EditorKenneth H. Rose, PMP

Contributing EditorLisa M. Fisher

PR O J E C T MA N A G E M E N T JO U R N A L

EDITORIAL REVIEW BOARDRick Bilbro, The Innova Group, Inc.;Tomas Blomquist, Umeå University, Sweden;Pierre Bonnal, CERN Accelerator and Beams Dept.; David Bonyuet, PMP, Delta Search Labs; Dan H. Cooprider, Creative People Management;Steven V. DelGrosso, PMP, IBM;J. Davidson Frame, University of Mgmt. & Technology;Ken Hartley, Parsons BrinckerhoffGary C. Humphreys, Humphreys & Associates;Peter Kapsales, Belcore; Young Hoon Kwak, The George Washington University; Alexander Laufer, Technion-Israel Inst. of Technology;William Leban, PMP, DeVry University; Ginger Levin, University of Wisconsin-PlattevilleRolf A. Lundin, Jonkoping University; Fred Manzer, PMP, Center for Systems Management;Peter Morris, University College, London; Ralf Müller, Umeå University, Sweden;Michael Okrent, Agilent Technologies Inc.;Blaize Horner Reich, Simon Fraser University.;Paul B. Robinson, Market Strategy and Analytics Partners;Richard L. Sinatra, PMP, Potomac, MD;James Snyder, Springfield, PA;Anders Soderholm, Umea University;Paul Solomon, PMP, B-2 Earned Value Management Systems;Charles J. Teplitz, University of San Diego;Janice Thomas, Athabasca University;Zeljko M. Torbica, PMP, Florida International University;Terry Williams, University of Strathclyde; Graham M. Winch, The University of Manchester.

EDITORIAL ADVISORY BOARDFrank T. Anbari, PMP, The George Washington University; Karlos Artto, Helsinki University of Technology(HUT), Finland; David Cleland, University of Pittsburgh;Lynn Crawford, University of Technology, Sydney; Brian Hobbs, Université du Québec à Montréal; Toshihiko Kinoshita, Waseda University; Mark E. Nissen, Naval Postgraduate School;Asbjørn Rolstadås, Norwegian University of Science and Technology; Dennis P. Slevin, University of Pittsburgh.

PUBLICATION & MEMBERSHIPThe Project Management Journal (ISSN 8756-9728/03)is published quarterly (March, June, September, December)by the Project Management Institute. The Journal is printedin the USA by Cadmus Magazine, Richmond, VA.Periodical postage paid at Newtown Square, PA 19073 USAand at additional mailing offices. Canadian agreement#40030957. Postmaster: Send address changes to ProjectManagement Institute, Headquarters, Four CampusBoulevard, Newtown Square, PA 19073-3299 USA; tel: +1-610-356-4600, fax: +1-610-356-4647.

© 2007 Project Management Institute, Inc. All rights reserved.

“PMI” is a trade and service mark registered in the United States and other nations; “PMP” and the PMP logo are registered certification marks in the United States and other nations;

“PMBOK” is a trademark registered in the United States and other nations; and the PMI logo, “PM Network”, “Project Management Journal”, “PMI Today”, and

“Building professionalism in project management.” are trademarks of the Project Management Institute, Inc.

PMI Regional Service Centre: Europe - Middle East- Africa (EMEA), 300, Avenue de TervuerenB-1150 Brussels, Belgium; tel. +32-2-743 15 73; fax +32-2-743 15 50; e-mail: [email protected]:[email protected]

PMI Asia Pacific Service Centre, #03-01, Rex House73 Bukit Timah Road; Singapore 229832; tel. + 65 63306733; fax. +65 6336 2263; E-mail: [email protected]:[email protected]

PMI Representative Offices, Beijing, ChinaWashington, D.C.

EDITORIAL & ADVERTISING SERVICESAddress manuscripts and other editorial submissions,advertising and mailing list rental inquiries, and requestsfor reprints/bulk copies/reprint permission to: ProjectManagement Institute, Publishing Department, FourCampus Boulevard, Newtown Square, PA 19073-3299USA. tel: +1-610-356-4600, fax: +1-610-356-4647; e-mail:[email protected]

Unless otherwise specified, all letters and articles sentto the PMI Publishing Department are assumed forpublication and become the copyright property of PMI ifpublished. PMI is not responsible for loss, damage, orinjury to unsolicited manuscripts or other material.

READER SERVICESPhotocopies. Project Management Journal has beenregistered with the Copyright Clearance Center, Inc.Consent is given for copying of articles for personal orinternal use, or for the personal use of specific clients.This consent is given on the condition that the copierpays through the Center the per copy fee stated in thecode on the first page of each article for copying beyondthat permitted by Sections 107 or 108 of the U.S.Copyright Law. The appropriate fee should be forwardedwith a copy of the first page of the article to theCopyright Clearance Center, Inc., 222 Rosewood Drive,Danvers, MA 01923 USA (tel: +1-508-750-8400; fax:+1-508-750-4744). The fee indicated on the first page ofan article in this issue will apply retroactively to allarticles published in the journal, regardless of the year ofpublication. This consent does not extend to other kindsof copying, such as for general distribution, resale,advertising and promotion purposes, or for creating newcollective works. Special written permission must beobtained from the publisher for such copying.

Permissions. Requests to reprint articles publishedin Project Management Journal must be made in writing tothe publisher.

Reprints. Individual articles may be purchasedthrough the Knowledge & Wisdom Center(www.pmi.org/k&wc)

Glossy Reprints. Requests for glossy reprints ofindividual articles in quantities of 100 or more can be sentto [email protected].

Bulk Copies of Current Issues. Copies of thecurrent Project Management Journal can be obtained inquantities of 25 or more. Orders must be placed 40 daysprior to date of issue. The cost is $5.50 per copy plusshipping.

Back Issues. Back issues are available on request at$20 each. Call +1-610-356-4600 for availability.

The mission of the Journal is to provide informationadvancing the state of the art of the knowledge of projectmanagement. The Journal is devoted to both theory andpractice in the field of project management. Authors areencouraged to submit original manuscripts that are derivedfrom research-oriented studies as well as practitioner casestudies. All articles in the Journal are the views of theauthors and are not necessarily those of PMI. Subscriptionrate for members is $14 U.S. per year and is included in theannual dues.

Claims for undelivered copies must be made no laterthan two months following month of publication. Thepublisher will supply missing copies when losses have beensustained in transit and when the reserve stock will permit.

PMI is a nonprofit professional organization whosemission is to serve the professional interests of itscollective membership by: advancing the state of the artin the leadership and practice of managing projects andprograms; fostering professionalism in the management ofprojects; and advocating acceptance of projectmanagement as a profession and discipline. Membershipin PMI is open to all at an annual dues of $119 U.S. Forinformation on PMI programs and membership, orto report change of address or problems with yoursubscription, contact:

Project Management Institute Global OperationsCenter, Four Campus Blvd, Newtown Square, PA 19073-3299 U.S.A.; tel: +1-610-356-4600; fax [email protected]; mailto: [email protected] Web site: www.pmi.org

3

F R O M T H E ED I T O RChristophe Bredillet, PhD, DSc, MBA, Ingénieur EC Lille

Exploring Research in Project Management—Nine

Schools of Project Management Research (Part 1)

In the next series of the “Letter From the Editor,” Iam going to present the results of research under-taken with my colleagues J. Rodney Turner1 and

Frank T. Anbari2 in order to draw a mapping of theresearches conducted in the field of project manage-ment and their main characteristics. The initial pur-pose was to get a better understanding of the kind ofresearch addressing project management, but indoing so, we raised our objectives.

We demonstrate that project management is agrowing field of study, of considerable diversity andrichness, worthy of full academic recognition, andsuggest that management will be richer for giving itsuch recognition. We review the substantial progressand trends of research in major schools of thought inproject management, address interactions betweenthese schools and with other related managementfields, and provide insights into current and poten-tial research in each and across these schools.

For the past 60 years, organizations have beenincreasingly using projects and programs to helpthem achieve their strategic objectives (Frame, 1994;Turner, 1999), while dealing with increasing com-plexity, uncertainty, and ambiguity affecting organi-zations and the socio-economic environment withinwhich they operate (Gareis, 2005). Through proj-ects, resources and competencies are mobilized tocreate competitive advantage and other sources ofvalue. Until the mid 1980s, interest in project man-agement was limited to engineering, construction,defense, and information technology. More recent-ly interest has been exploding. Currently, at leastone fifth of global economic activity takes place asprojects. Developing relevant competence at all lev-els, individual, team, organizational, and society isseen as a key for better performance (Gareis &Huemann, 2007). Educational programs andresearch in project management have been alsogrowing rapidly during the last three decades to sup-port the need for competence and the underlyingsupporting knowledge field (Atkinson, 2006;Umpleby & Anbari, 2004).

Organizations, both public and private, havebeen undertaking projects for millennia, but whatwe understand as modern project and programmanagement has only been used and developedsince the late 1940s, driven initially by the econom-ic development after World War II and the arms race(Morris, 1997). Modern project management startedas an offshoot of operations research, with the adop-tion of optimization tools developed in that field,but has subsequently broadened so that at least nineschools of thought in project management can nowbe identified. However, to support the current devel-opment of project management, the need is growingamong the management research community andprofessional associations to clarify project manage-ment research trends to support the progress of suchresearch and the development of competencies assources of performance improvement and value cre-ation according to the current socioeconomic con-text and various management situationsorganizations have to face (Bredillet, 2005).

To support these developments, the projectmanagement research community needs to be a fulland recognized part of the academic community inmanagement, so that academics in the subject canreceive full recognition for their work and others areencouraged to pursue research in related areas.However, project management is generally ignoredas an academic subject by most of the managementresearch community. Evidence of this includes:• There is no division at the Academy of

Management explicitly addressing project man-agement (although at the annual conference of2007, for the second time ever, a professionaldevelopment workshop in project managementis planned).

• The Academy of Management Journal has publishedonly four papers on project management topics inthe last 20 years (Conlon & Garland, 1993;Harrison & Harrell, 1993; Lewis, Welsh, Dehler, &Green, 2002; Yakura, 2002).

• The Academy of Management Review has not pub-lished a single paper on project management top-ics in the last 20 years.

JU N E 2007 PR O J E C T MA N A G E M E N T JO U R N A L

1 ESC Lille—Lille School of Management, Lille, France

2 The George Washington University, Washington, DC, USA

44

• The Journal of Management does not have project man-agement as one of its subject areas, although it does haveoperations management and technology management.

There are oases of light. The European Academy ofManagement (EURAM) has had a track on project man-agement at all seven of its conferences starting in 2001. Itis the only subject to have achieved that status and it hasbegun to attract the most papers and participants at theconference. It is ironic though, that project managementshould be treated with such disdain by the rest of the aca-demic community in management. Fifty years ago, man-agement had to fight for recognition by the rest of theacademic community, and has since deservedly achievedthat. We feel the management research communitywould be more enriched if it gave greater recognition toproject management.

World Bank (2005) data indicate that 21% of theworld’s gross domestic product (GDP) is gross capitalformation, which is almost entirely project-based. Inmany public and private organizations, some operat-ing expenditure is also project-based. Thus, the aca-demic management community is generally ignoring afield of management that addresses how more thanone fifth of value generation in the world can best bemanaged, and is therefore failing to properly supportpracticing managers. The academic community, thepractice of management, and the world economywould be more enriched if project management weretaken more seriously.

We have written this with the aim of showing thatproject management is a field of diversity, complexity,and richness worthy of academic study and research, witha contribution to make to the study of management, andworthy of recognition by the rest of the academic com-munity in management. To this end we address the fol-lowing objectives:1. To demonstrate why project management is a recogniz-

able field of study worthy of full academic recognition.2. To identify the diversity and richness of the subject as

evidenced by at least nine schools of thought.3. To identify how project management continues to

draw on and make contributions to other fields ofmanagement, and therefore explain how managementwill be more enriched for giving it full recognition.

We based our paper on an extensive, careful review ofacademic research literature on project management thatreflects the evidence advanced by leading thinkers andresearchers in the field. We organized the literature into ninemajor schools of thought on the basis of the key theoretical

idea that drives each one. Our intent of separating theseschools of thought is to gain insight into current and poten-tial research, within a manageable number of researchthemes without over-simplification of the richness of theunderlying thought. However, we also discuss the overlapand interactions between project management schools.

In the next “Letter From the Editor,” I will defineproject management as being a recognizable field ofstudy. Before introducing the different schools of thoughtwe discovered and their characteristics in future Editions.

Ordo ab chaosChristophe N. Bredillet

References

Atkinson, R. (2006). Guest editorial: Excellence inteaching and learning for project management.International Journal of Project Management, 24, 185–186.

Bredillet, C. N. (2005). Editorial: The good, the bad,and the ugly. Project Management Journal, 36(1), 3–4.

Conlon, D. E., & Garland, H. (1993). The role ofproject completion information in resource allocationdecisions. Academy of Management Journal, 36, 402–413.

Frame J. D. (1994). The new project management: Toolsfor an age of rapid change, corporate reengineering, and otherbusiness realities. San Francisco: Jossey-Bass.

Gareis, R. (2005). Happy projects! Vienna, Austria: Manz.Gareis, R., & Huemann, M. (in press). Maturity mod-

els for the project oriented company. In J. R. Turner(Ed.), The Gower handbook of project management.Aldershot: Gower.

Harrison, P. D., & Harrell, A. (1993). Impact of“adverse selection” on managers’ project evaluation deci-sions. Academy of Management Journal, 36, 635–643.

Lewis, M. W., Welsh, M. A., Dehler, G. E., & Green, S.G. (2002). Product development tensions: Exploring con-trasting styles of project management. Academy ofManagement Journal, 45, 546–564.

Morris, P. W. G. (1997). The management of projects(2nd ed.). London: Thomas Telford.

Turner, J. R. (1999). The handbook of project basedmanagement (2nd ed.). London: McGraw-Hill.

Umpleby, S., & Anbari, F. T. (2004). Strengthening theglobal university system and enhancing projects manage-ment education. Review of Business Research, 4, 237–243.

World Bank. (2005). Little Data Book. Washington,DC: International Bank for Reconstruction andDevelopment. The World Bank Development Data Group.

Yakura, E. K. (2002). Charting time: Timelines astemporal boundary objects. Academy of ManagementJournal, 45, 956–970.


5

MANAGING KNOWLEDGE AND LEARNING INIT PROJECTS: A CONCEPTUAL FRAMEWORKAND GUIDELINES FOR PRACTICE

This paper presents a framework identify-

ing the key areas within IT projects where

knowledge-based risks occur. These risks

include a failure to learn from past proj-

ects, competence of the project team,

problems in integrating and transferring

knowledge, lack of a knowledge map, and

volatility in governance. The model was

compiled through an extensive literature

search encompassing project manage-

ment, information systems, software

development, and team learning litera-

tures. This framework was then tested

and modified through a field study of 15

senior project managers from North

America and New Zealand. Analysis of the

interviews from the field study resulted in

a set of five broad principles of knowl-

edge management within projects. These

principles relate to a climate for learning,

knowledge levels, knowledge channels,

team memory, and knowledge risks.

Practices suggested by the interviewees

accompany each principle.

Keywords: IT projects; knowledge-based

risk; risk models

©2007 by the Project Management Institute Vol. 38, No. 2, 5-17, ISSN 8756-9728/03

Introduction

History has shown that it is difficult to demonstrate the value generated fromorganizational spending on IT projects (Brynolfsson & Hitt, 1998; Heller,2000). According to a Standish Group study, IT projects have a 66% failure

rate; such projects have either missed the targets or failed to deliver the requiredbusiness functionality (Standish Group, 2003).

Other reports show that the rate of project success is increasing, caused per-haps by increased usage of project management methods and by a heightened con-centration on controlling project size and scope (Sauer, Gemino, & Reich, 2006).However, discussions with project managers have indicated that IT projects arebecoming more ambitious, more organizationally and technically complex, andmore time-to-market focused. Acceptable project performance may still be an elu-sive target (Sauer & Reich, 2006).

Increasingly, IT projects are relied on by organizations that are trying to inno-vate or to respond to competitive threats. Transformations such as mergers, supplychain integration, and customer service innovation are necessary for organizationsto survive and prosper. Therefore, the high failure rate of IT projects is a majorstumbling block to companies that are trying to innovate through new processesand services.

For years, practitioners and researchers have repeated the same advice to com-panies trying to improve their track record with IT projects (e.g., Standish Group,2003): increase executive sponsor support, reduce scope and duration, employproject management methodologies, and manage change requests. Although con-tingency approaches have been suggested (Barki, Rivard, & Talbot, 2001), theadvice has remained largely unchanged.

One promising lens through which projects can be understood and poten-tially improved is to conceptualize them as arenas in which knowledge is generat-ed and exploited and learning is essential for success. The foundations for thisapproach are found both in the knowledge management and organizational learn-ing literatures. In addition to this literature, which mainly focuses on permanentorganizations and teams, many project-oriented researchers have been investigat-ing aspects of learning and knowledge management in projects (e.g., Bredillet,2004; Faraj & Sproull, 2000; Tiwana, Bharadwaj, & Sambamurthy, 2003; Walz,Elam, & Curtis, 1993).

BLAIZE HORNER REICH, Simon Fraser University, Canada

ABSTRACT

5JU N E 2007 PR O J E C T MA N A G E M E N T JO U R N A L

66

The author and colleagues have been investigating proj-ects using the knowledge management lens for three yearsin a multi-stage research project. This research has twobroad goals:

1. To create a framework which comprehensively maps theknowledge management research relating to IT projects.

2. To provide practical guidelines and practices for eachknowledge-based risk identified in the model.

The research community is the target of the first goaland the practicing project manager and project office is thetarget of the second.

The paper is organized into six sections. After this intro-duction, a section on the origins of knowledge managementand learning is presented. The research methodology andthe conceptual framework, which includes a definition, aknowledge typology and a model comprising 10 knowledge-based risks in IT projects are then presented. The last twosections present guidelines and principles to mitigate theserisks, summarizes and suggests further research.

Knowledge and Learning Within a Project

Knowledge has a long established role in theories designed toexplain organizations and their performance. In earlier theo-ries it appeared under the guise of “technology.” Lately, theo-rists have given knowledge a more explicit role of its own.This focus on knowledge derives from several key insights:

• Environmental turbulence (e.g., regulatory and technol-ogy changes) has forced many organizations to innovatefaster and to operate more efficiently. This pressurerequires a focus on continual learning and renewal.

• Specialist or distinctive knowledge embodied in prod-ucts or services can be a source of competitive advan-tage and therefore drive value creation.

• The means by which organization is effected (lines ofauthority, integrating committees, etc.) can be conceptu-alized in terms of knowledge and information processing.

• Knowledge and information have characteristics thatdistinguish them from other organizational resources,e.g., persistence, negligible cost of duplication, lowvisibility, and intangibility.

Two important consequences emerge:1. Many, if not all, organizational outputs can be use-

fully seen in knowledge-based terms, so we need touse knowledge-based concepts and theories to under-stand how they are produced.

2. The structure and processes of organizations and thepeople and systems within them can be usefullyunderstood in knowledge-based terms.

There have been two streams of research that haveresponded: organizational learning and knowledge manage-ment. The organizational learning stream, beginning withCangelosi and Dill (1965), was fueled by seminal works inthe 1970s (e.g., Argyris & Schön, 1978; March & Olsen, 1975).There was a resurgence of interest in this area beginning in the

early 1990s (Brown & Duguid, 1991; Huber, 1991; Pedlar,Burgoyne, & Boydell, 1991; Senge, 1990; Walsh & Ungson,1991; Weick, 1991), which has remained to this day.

Organizational learning research extends the theory ofindividual learning to the organizational context, seekingto understand how teams and groups learn and why someorganizations are better at learning than others. Thisresearch has brought us concepts such as single- and dou-ble-loop learning, organizational memory and forgetting,and system dynamics. The focus of this research is mainlyin permanent teams and organizations.

Knowledge management research, after seminal workby Polyani (1958, 1966) began in earnest in the 1990s (e.g.,Cohen & Levinthal, 1990; Nonaka, 1991; Nonoka &Konno, 1998; Wenger, 1998; Zack, 1999). It seeks toexplain how knowledge is acquired, transformed, and even-tually exploited. One branch of knowledge managementresearch focuses on technologies to support knowledgemanagement (e.g., Alavi & Leidner, 1999, 2001). The main-stream theoretical research has developed and studied con-cepts such as tacit and explicit knowledge, knowledgecreation, individual and organizational knowledge, andcommunities of practice. Much of the focus in this researchis on the tacit knowledge that is embedded in individualsand the study of how this knowledge can be integratedwith other knowledge, shared with other individuals, andeventually made explicit so it can be aggregated and distributed widely.

Projects, whether or not we choose to think of them astemporary organizations (Lundin & Soderholm, 1995;Packendorff, 1995; Söderlund, 2002; Turner and Muller,2003) involve considerable knowledge processing. In trans-formational projects, teams of individuals from diverseorganizations with different specialist knowledge worktogether under time and budget constraints to produce anew product, process, or service. From this perspective, onemight say that a project manager’s primary task is to managethe knowledge bases of the team members and stakeholdersso that they combine in the best possible way to successful-ly accomplish their assignment. Team members individual-ly and collectively have to learn new knowledge, transfertheir knowledge to others, and create new shared under-standings at the right times and for the right cost. Therefore,the concepts of individual and project-based learning arevery relevant to the success of these projects. Concepts suchas knowledge creation and dissemination are critical as well.

Several researchers have questioned the connectionbetween projects and knowledge or learning. Thomas andTjader (2000) identified a disconnect between the projectas a field of action and a field of reflection. However,Leseure and Brookes (2004), who started with the hypoth-esis that project management and knowledge managementare conflicting paradigms, reached the conclusion that “proj-ect management and knowledge management can only gohand-in-hand.”

In addition to conceptual support for the importance ofknowledge and learning in projects, three empirical studies


77

from the IT literature have shown that managing knowledgecan add to the probability of project success:

• A survey of project team members and stakeholdersworking on 69 software development projects (Faraj& Sproull, 2000) found that by coordinating expert-ise, teams could improve their performance by 25%more than by applying traditional project manage-ment practices.

• Yoo and Kanawattanachai (2001) tracked the progressof 38 virtual project teams and found that project suc-cess was strongly influenced by each team member’sknowledge of other team member’s areas of expertiseand by the team’s ability to harness this knowledge toachieve the project’s goals.

• A study of 133 projects (Tiwana, Bharadwaj, &Sambamurthy, 2003) reported that the level of knowl-edge integration within the team was more importantto project success than a pre-existing positive relation-ship between business and IT.

Therefore, the study of learning and knowledgemanagement within IT projects presents a fruitfulavenue of research. To pursue this course, we can takeone of two approaches. Either we attempt to apply firm-level and organizational-level concepts to the project, orwe start with the project and attempt to build immedi-ately relevant concepts for addressing its knowledge-based character.

An analysis of these two approaches resulted in thefollowing conclusions:

Work has already begun in several research disci-plines to connect projects with the learning and knowl-edge concepts, from the “project-up” perspective. Forexample, research has begun into how project membersand organizations could learn within and across projects(e.g., De Fillippi, 2001; Keegan & Turner, 2001; Kotnour,1999; Schindler & Eppler, 2003).

We have completed related research to examine thecurrent professional practice standards of managingknowledge in projects. Findings suggested that projectmanagers lack the guidelines needed to successfullymanage knowledge-based risk (Reich & Wee, 2006).

The gap between the highly conceptual organiza-tional learning and knowledge management literaturesand the more practice-oriented and largely non-theoret-ical project management literatures is very wide.


There might be a way to fruitfully bridge this gap, but itwas decided to start at the level of the project, to integrate andmodel the work that had already been begun at this level.

Therefore, we began this research with a thorough exami-nation of the existing literature that ties projects together withconcepts of learning and/or knowledge management. The con-ceptual framework was developed and then “tested” with anextensive field study. The field study suggested principles andpractices that may have immediate relevance to practicingproject managers. Our goal is to create a holistic view of theways that knowledge impacts the outcome of IT projects.

Methodology

This research used a methodology comprised of five major steps.These steps are summarized in Table 1 and discussed next.

Step 1: Literature SearchThe literature on projects, especially IT projects, wasreviewed, using a three-step approach:

1. Literature search using keywords such as projectmanagement, project, or software development inconjunction with knowledge and/or learning

2. Expansion of the search to include the literature onorganizational behavior and the keyword list toinclude terms such as teams and learning

3. Selection of articles that discussed learning or knowl-edge within a project or team environment publisheduntil mid-2006.

More than 50 articles were found by using thisapproach. These articles emanated from four distinct andseparate research disciplines: project management, learningin teams, software engineering, and management informa-tion systems. A working paper format was used to analyzeand integrate these articles (Reich, 2006).

Step 2: Model BuildingThere are several approaches to aggregate literature in orderto reveal the overlaps between disciplines as well as developand test common concepts and assumptions. Rosemannand Chan (2000) suggested three dimensions with which toexamine knowledge management in projects: the systemlifecycle, the knowledge lifecycle, and the knowledge con-tent taxonomy. This research adopts the first option—thesystem lifecycle—and generalizes it using the model out-lined in Arthur, De Fillippi, and Jones (2001).

Table 1: Steps in research methodology

88

This approach resulted in a model, originally called theKnowledge Traps Model, which mirrors the input-process-out-put model commonly discussed in information systemsdevelopment literature. The model graphically depicts theplaces in an IT project where knowledge and learning issuescould impact IT project success. These knowledge risks arecategorized as risks of project inputs, governance, projectprocess, or project outputs. In each of these parts of a project,knowledge may be unavailable, lost, or simply not created.

Step 3: Taking the Model on Tour After completing the initial model, the author took it on tourto gauge the interest it generated among academics and to testits salience with project managers. She spoke at universities,conferences (e.g., PICMET [Portland International Center forManagement of Engineering and Technology] and IPMA[International Project Management Association]) and to PMIand CIO audiences in Canada and New Zealand. These dis-cussions generated ideas that were added to the model.

In general, support for the insights contained in themodel was strong, with many project managers suggestingthat they routinely faced knowledge-based risks but lacked aformal process for acknowledging or mitigating these risks.

Step 4: Data Gathering Using the initial Knowledge Trap Model, the author designedan interview guideline and assembled a sample of seniorproject managers for interviews. The search for interview par-ticipants was focused on finding project managers who wereinterested in knowledge and learning and who had devel-oped practices around these topics. Some participants werereferred by initial recruits and a call was sent out to severalPMI chapters that the author had worked with. The interviewprocess was terminated when few new perspectives or prac-tices were being discussed by the interviewees.

In total, 15 senior project managers were interviewed.These individuals—working in either North America or NewZealand—were employed by a variety of organizations,including for-profit organizations and government depart-ments. Some were in-house project managers; others, con-sultants. The interview focused on four areas:

1. Interviewees’ unprompted ideas about knowledgemanagement within a project.

2. Their experiences in resolving each of the 10 knowl-edge traps.

3. Their suggestions for making additions or changes tothe model.

4. Other knowledge-based challenges they have confront-ed in managing each of the project lifecycle stages.

Step 5: AnalysisUsing atlas.ti software, the interviews were catalogued,cleansed, and qualitatively analyzed. The data confirmedthat the knowledge traps did indeed exist, and that each oneposed different levels of risk for project managers. It wasalso possible to create a typology of knowledge that wasimportant to project managers.

The project managers told stories about their knowl-edge and learning experiences in projects. From these, it waspossible to extract common concepts and solutions thatthey deployed and to aggregate these into principles forpractitioners.

The next section presents the conceptual frameworkdeveloped from this literature and field research.

A Conceptual Framework of Knowledge and Learning

in IT Projects

The conceptual framework contains three parts: • A definition of knowledge management in the context

of IT projects• A typology of knowledge that is critical to IT project

success• A model that identifies the knowledge-based risks in

an IT project.

Each of these elements is presented next, along with thesupporting research.

What is Knowledge Management in IT Projects? Interviews with project managers identified a lack of com-mon understanding about the meaning of knowledge man-agement within a project context. Some intervieweesfocused on explicit knowledge and various ways to developand deploy artifacts such as project websites and commonfolders. Many mentioned “lessons learned” as the mostimportant knowledge concept. Others understood knowl-edge management as a more complex concept, ranging fromefficiently using experts’ tacit knowledge to actively plan-ning a project team’s knowledge needs.

From the interviews, there was evidence to support theidea that project managers do consider their team as anaccumulation of knowledge capital (Arthur et al., 2001), butthe participants reported difficulty in harnessing this capitalbecause they lacked a systematic knowledge-based method-ology. From research in both knowledge management andproject management, the following preliminary definition isproposed:

Knowledge management in the context of a project is theapplication of principles and processes designed to make rel-evant knowledge available to the project team. Effectiveknowledge management facilitates the creation and integra-tion of knowledge, minimizes knowledge losses, and fillsknowledge gaps throughout the duration of the project.

The next section outlines the knowledge-based risksand uses quotes from the interviewees to illustrate the issuesthey face.

What Knowledge is Important to IT Projects?

This project’s research suggests that there are four types ofknowledge important to the success of IT projects: process,domain, institutional, and cultural.

The literature on learning and knowledge managementin projects suggested the importance of two types of knowl-


99

edge—process and domain (Reich, 2006). The first type,process knowledge is the knowledge that team members andsponsors have about the project structure, methodology,tasks, and time frames (Chan & Rosemann, 2001; Meehan& Richardson, 2002). This knowledge allows a team mem-ber to understand his or her part in the overall project andto understand what is expected and when it is to be deliv-ered. This kind of knowledge also allows a team or subteamto self-organize, since the team knows the outputs requiredand the time frames and can, if empowered, decide how thework should best be accomplished.

The other kind of knowledge suggested by many researchstudies is domain knowledge—the knowledge of the industry,firm, current situation, problem/opportunity, and potentialsolutions (including technology and process). This knowledgeencompasses three types identified in Chan and Rosemann(2001): business, technical, and product knowledge.

This knowledge is spread widely within and outsidethe project team. The project sponsor may be the mostknowledgeable about the industry and the problem oropportunity being addressed. Technical experts inside andoutside the company have knowledge about the technolo-gies that could be brought to bear. Project members willhave deep knowledge about the company and its businessprocesses. Many researchers have discussed domain knowl-edge as it relates to the issues of knowledge integration(e.g., Nelson & Cooprider, 1996; Majchrzak & Beath, 2003;Walz et al., 1993); coordination (Crowston & Kammerer,1998; Faraj & Sproull, 2000); learning (Stein &Vandenbosch, 1996); and governance of projects (Henry,Kirsch, & Sambamurthy, 2003).

One other kind of knowledge essential to project suc-cess was suggested by attendees at the talks delivered to prac-ticing project managers. There were various names given toit in these discussions, but the name proposed for this typol-ogy is institutional knowledge. This knowledge is a mix of thehistory, the power structure, and the values of the organiza-tion. This knowledge will be transferred by means of storiesor anecdotes by insiders and observers of an organization. Itis not as much about facts as it is about how the facts can beinterpreted to understand “what is really going on.” Thisknowledge, we heard, was particularly important for anexternal project manager or a vendor to access in order to getdifficult problems dealt with and key decisions made in thecourse of a project.

The fourth kind of knowledge necessary in an IT projectwas suggested recently by Dr. Rodney Turner in conversa-tion. He suggested that cultural knowledge was particularlyimportant in IT projects. He noted that a project managerneeded to understand how to manage IT people, who arethought to have fairly unique cultural norms. However, in abroader context, with project teams being comprised ofmany disciplinary groups (e.g., web designers, organization-al development experts, IT architects) and people frommany cultural backgrounds, the concept that cultural knowl-edge, both discipline-based and national, might be impor-tant is a very useful idea.

In summary, there are many types of knowledgeneeded within a single project. The more complex andinnovative the organization, the problem or opportunity,or the technology within the project, the more importantit will be to mobilize and exploit these different types ofknowledge.

Knowledge-Based Risks in IT Projects

Many sources were explored to understand how academicswere conceptualizing knowledge and learning in projects.One finding was that the approaches taken and the con-structs discussed in the project management, software devel-opment, management information systems, and teamlearning literature is diverse and fragmented (Reich, 2006).However, rather than trying to unravel the differences andknit together the similarities of this literature, we have creat-ed an integrated model. Our interest was to understand at ahigh level what was being said and to subject this integratedview to the scrutiny of researchers and practitioners.

We originally used the term Knowledge Trap to identifythose times or events within an IT project in which there is aloss of project-specific knowledge, where the project lackssome relevant knowledge, or where knowledge is not createdor applied optimally (Reich, 2004). As the research progressed,however, we understood better that these issues of knowledgeand learning could be thought of as risks to the project, andtherefore we have recast the model in risk-based terms.

Interviews confirmed that many accomplished projectmanagers know intuitively about the knowledge risksdetailed in this paper. One participant explained it thisway: “I would say we saw all kind(s) of instances whereknowledge caused challenges, and the challenges typical-ly manifested themselves in a piece of functionality beingdeveloped, which totally missed the mark… at the techni-cal level, we’d … develop a piece of code which basicallydidn’t work.” Another project manager noted that knowl-edge issues had caused problems on almost every singleproject she had worked on.

It was difficult to get respondents to quantify theimpact of knowledge risks because many had not explicit-ly considered this impact before the interview. One inter-viewee, however, estimated that in his recent (US)$60million project, knowledge issues added about 10% to thefinal cost.

Figure 1 illustrates the four parts of the Knowledge Risks model.

1. Inputs: A project’s knowledge inputs 2. Process: A project’s governance 3. Process: A project’s operational phases—plan,

design, build/configuration, and implementation 4. Outputs: A project’s delivery and its closeout.

Inputs

There are two main knowledge risks at the beginning of aproject: the failure to learn from past projects as well asthe failure to meet the project’s knowledge needs duringteam selection.


1010

Risk #1: Lessons not Learned Project teams often contain individuals from many differentorganizations and units. These professionals may or maynot have attempted a project similar to the one they areabout to start. Without access to lessons learned from com-parable projects, the team will lack both important knowl-edge of the project’s risks and the opportunity to shareinformation about the ways these risks could impact theproject’s plan and targets.

Although Kotnour (1999) concluded that lessonslearned are conducive to learning within and across projects,this research confirms other research showing that manyorganizations continue to fail to learn from projects(Thomas & Tjader, 2000).

From informal discussions as the model was present-ed and from the qualitative analysis of the interviews, itwas very clear that few projects used the lessons learnedfrom prior projects. Most participants said somethingsimilar to “We normally are so anxious about getting thething going that we just do whatever seems to come tohand.” Because everyone felt pressure imposed by timeconstraints, they put their emphasis on go-forward activ-ities and did not take the time to look reflectively backinto their shared history. However, when asked about thepotential impact of learning from past projects, one

interviewee suggested that his most recent project wouldhave seen “a 10 to 15% change in the results.”

Risk #2: Team Selection Flawed Finding the right team members is important because theproject manager will want all relevant knowledge areasincluded in the team or available to the team when theyneed it. Walz et al. (1993) concluded that “assigning oneor two individuals with deep application domain andtechnical knowledge to a design project can significantlyreduce the learning time involved.” Grant (2006) provid-ed specific guidelines to help in the selection of theknowledge experts.

Of the project managers interviewed, all reported think-ing about the knowledge and the skills they needed for eachproject. Some had very systematic ways to document theseneeds. They noted, however, that problems arose when theylacked influence over team selection. Also, knowledge pro-files of team members were often not available during theselection process. One senior project manager reported thathe knew what he needed, but “identifying the resources tomeet those skill sets was a whole different ballgame.”

When the team selection process is flawed, the projectmanager will not know what the team knows collectively or,more importantly, what it doesn’t know.


Figure 1: Knowledge-based risks in IT projects

1111

Governance Processes

Project governance is the responsibility of several key posi-tions: executive sponsor, project champion, project manag-er, and project steering committee. From a knowledgeperspective, effective project governance involves twoissues—volatility and role understanding.

Risk #3: Volatility in the Governance Team After a project begins and the governance structures are putin place, there is a gradual knowledge-building processamong the key stakeholders. This knowledge risk refers tothe loss of any member of the governance structure whocontrols or influences project resources and direction.Whether moving a project manager or other executiveaway from a project is done for strategic reasons orwhether it is totally unplanned, there is a knowledge lossthat may cause targets to be missed and benefits to gounrealized. Empirical research shows that a change of proj-ect manager is the most important governance volatilityissue affecting project performance (Parker & Skitmore,2005; Sauer et al., 2006).

One respondent estimated that on his most recentproject, losing the project sponsor “probably delayed theproject by a year.” Another respondent, commenting on theloss of a project manager, said, “If we had given whateverwe could, reasonably, to bring this person back, even half ayear’s salary or bonuses or whatever, it would have beenworthwhile. At the end of the day, the cost, I believe, is pro-portionate to the stakes that we were managing.”

Risk #4: Lack of Role Knowledge Among the Governance Team Gaps in project-governance knowledge endanger the suc-cess of projects. When senior executives assume the roles ofproject sponsor or project champion, they do so becausethe outcome of the project is important—to them and totheir organization. Although many researchers acknowl-edge that there is a difference between managing the staticnature of business operations and the dynamic state ofprojects, organizations frequently neglect to provide theirexecutives with the training they need to function as proj-ect sponsors and project champions (Henry et al., 2003).

A first-time project sponsor may not know when tosupport the project and when to tighten up the reins, ormay fail to differentiate between those project problemsthat are serious and those that are temporary. One intervie-wee recognized this by noting, “They don’t know [that] …ifthey do not make a decision, they are wasting a hundredthousand dollars [as] things churn for another week.”

One of the key problems that emerged from the inter-views was the difficulty project managers have whenattempting to “educate” a project sponsor, since they pos-sess very different levels of authority and power within theorganization. A related issue may well be the difficulty aproject sponsor has in admitting that he or she needs addi-tional training or that he or she does not know how to sup-port a project.

Operational Processes—Plan, Design, Build, and Implement

Within the main body of an IT project are five knowledgerisks: knowledge integration, knowledge transfer, loss ofteam members, lack of a knowledge map, and lossbetween phases.

Risk #5: Inadequate Knowledge IntegrationKnowledge integration is the process of bringing differentspecialist forms of knowledge (i.e., cross-functional) togeth-er to address an identified issue and to create knowledgethat is greater than the sum of its parts: a new idea, a sharedunderstanding, or an integrative model. Knowledge integra-tion is particularly important within a project that seeks totransform, create, or solve difficult problems.

Researchers have talked of the need to integrate theknowledge between the technical and business team mem-bers (Garrety, Robertson, & Badham, 2004) and examinedthe issues that project managers face when integratingknowledge in these cross-functional team environments(Huang & Newell, 2003). Other research (Postrel, 2002;Tiwana, 2003) has tried to determine how much knowledgeneeds to be integrated for optimum results.

In interviews, participants confirmed that for a projectthat is building a custom application, knowledge integra-tion between users and analysts as well as between teammembers is critical for achieving a successful outcome.Interviewees talked about the many dimensions involved inthis issue. One noted a project team problem: “How theteam feeds information amongst themselves. Especially in ahuge project in a larger organization, invariably what hap-pens is one team over here starts to solve a problem thatanother team is already solving.”

Risk #6: Incomplete Knowledge TransferWithin a project that is implementing packaged softwareor new hardware, knowledge transfer from vendor or con-sultant to the internal project members is critical. Manyresearchers have explored this area and documented theirfindings in the wider project and innovation literatures(Boh, 2003; Sense, 2003; Volkoff, 2001; Walz et al., 1993).

One project manager noted the challenge of integra-tion at his level: “It’s all about the transfer of knowledgefrom them to me. If they can’t illustrate, in some way, howthat works, then I can’t get it in my mind and make deci-sions that are good for the company.”

The project managers interviewed noted a conflict ofinterest that results when clients ask vendors to transferknowledge. For most vendors, their intellectual propertyrepresents a future revenue stream and they are reluctantto sell it, even if they could determine a price. One par-ticipant explained it this way: “The last project we workedon, one of the things that needed to be planned for wascontinuity, in terms of what happens when the contract-ed services come to an end. If the incumbent vendor isnot reappointed, then how do they transfer knowledge toa competitor—that’s always a touchy subject for an out-going vendor.”


1212

In general, the participants recognized misalignedgoals as a troubling project problem. Although as projectmanagers they could see both sides of the issue, theyoften needed to obtain more control over the vendor’sintellectual property in order to fully understand andmanage the project.

Another issue they noted involved measuring theknowledge transfer. Although project managers may includein the project plan activities to support knowledge transfer,they often have no objective way to measure how effectivelythese activities were performed.

Risk #7: Exit of Team MembersResearchers have noted the loss of knowledge when teammembers leave (Gable, Scott, & Davenport, 1998; Eskerod &Blichfeldt, 2005). Schindler and Eppler (2003) also notedthe cost to clients when critical external consulting resourcesleft a project.

Although project managers conceptually understandthat losing key team members results in knowledge gaps,they often fail to create a plan to mitigate such losses. Asone respondent noted about her planning: “It’s very reac-tive, and when somebody is leaving, then you kind of puta plan in place to keep him a little bit longer, or coachsomebody, or whatever.”

Experience teaches that the longer the project, thegreater is the risk that key people will leave through plannedand unplanned events. One participant noted that on a proj-ect lasting more than three years, he lost more than 100% ofthe project team complement. “We tried to reduce theimpact for scheduled moves, we tried to do a shadow activ-ity for a few weeks, or a few days, or sometimes no days. Theless time we had to shadow, the worse the outcome was.”

Risk #8: Lack of a Knowledge Map IT project managers and team members make a multitude ofinterrelated decisions. The more difficult the domain prob-lems are, the more important it is that team members pos-sess a knowledge map—information on the knowledgewithin the team (i.e., who knows what) and the knowledgeavailable to the team—that enables them to address com-plex problems efficiently and effectively.

Grant (2006) recommended the creation of a knowl-edge map to support the team’s ability to access each others’expertise. Von Stamm (2005) stated “networks are impor-tant for innovative projects because it is often not possibleto identify what kind of skills will be needed in the courseof the project at the outset.”

In their qualitative study, Crowston and Kammerer(1998) studied a group of requirements engineers to findout how they created requirements that were complete,accurate, and consistent. By using theories of transactivememory (Wegner, 1987) and collective mind (Weick &Roberts, 1993), they were able to explain their findings.Faraj and Sproull (2000) provided quantitative supportfor Crowston and Kammerer’s findings using expertisecoordination theory. These studies suggest that creating a

knowledge map and a knowledge network within a projectteam may prove more effective than focusing on knowl-edge integration.

Our interviewees agreed that this knowledge trap wasimportant. One project manager said that he needed amethodology “that would identify who knows what andmake a mental map of the team.”

Risk #9: Loss Between PhasesBecause team composition often changes from phase tophase in IT projects, there is a significant risk that the knowl-edge generated by one phase will be inadequately transmit-ted to the next phase. Traditional methods ofdocumentation—models, state diagrams, and use cases—rarely capture the “why” of design choices. As the next teaminterprets these artifacts, they may introduce errors into theproject or slow it down trying to understand the meaning ofprevious decisions. This problem is exacerbated in longprojects and within virtual project teams (Rus, Lindvall, &Sinha, 2001).

As one project manager noted: “There is always the archfrom analysis to design. A touch of magic, and this analysisbecomes that design. I just … had this big confrontationabout how the analyst thought that the people were goingoff and designing something that has nothing to do with therequirements, and the designers thought that the analystsweren’t talking to them.”

Another noted the issue of between-team understand-ing: “You create all these specifications up front…but speci-fication is language. I have to translate it into words andthen some other dude is going read my words and is goingto say, yes I agree with you. And then we go away, and we aredeveloping and then …after a year we come back togetherand of course it is not the same interpretation.”

Project Outputs

The most important knowledge risk at the end of the projectis one that many participants said topped their list—that les-sons learned are rarely satisfactorily captured.

Risk #10: Failure to LearnAs organizations become more reliant on projects for trans-formation and renewal, the outcome of a particular projectmay be less important that an overall increase in the abilityof an organization to implement projects successfully. Anincomplete debriefing during and at the end of a projectleaves team members with a fragmented idea of what waslearned and why things went wrong or right. When projectmanagers fail to capture lessons learned, they prevent team-level learning and hinder opportunities to improve organi-zational competency in managing and completing projects.

Middleton (1967) identified that lessons learned in oneproject did not find their way into the next projects. Almost40 years later, the situation has not improved significantly.Williams (2004) noted that the practice of post-mortemsdid not occur frequently. In a further study, Williams (2006)noted that there is little guidance on how to do project


1313

reviews, and that only 12% of surveyed participants saidthat procedures were followed. He concluded that the trans-fer of lessons learned to the organization from past projectsis the least successful aspect of learning.

Other than time pressure identified by Keegan and Turner(2001), Williams (2006) and participants in this research iden-tified cultural reasons as impediments to the lessons-learnedactivity. One project manager explained it this way: “I’ve neverworked in an environment where they are interested [in les-sons learned] or where it is a safe environment.”

Knowledge Principles and Practices

Projects will differ in their need for knowledge-based riskmanagement. Projects in which (1) the future state of theorganization is clearly documented, (2) a low level of inno-vation is required, and (3) there will be a stable governanceand project team, may be successful by focusing on tradi-tional project management—managing scope, time, andcost. However, when the organization or project is undergo-ing significant change, principles and practices of knowl-edge management become important.

Through an examination of the research literatureand interviews with the participating project managers,five broad principles of knowledge management withinprojects have been identified. These principles relate to aclimate for learning, knowledge levels, knowledge chan-nels, team memory, and knowledge risks. Within eachprinciple, the informants suggested practices for imple-mentation. Each principle and its associated practices arediscussed next.

Establish a Learning Climate

De Souza (2003) warned against an over-reliance on tech-nology without building a climate that supports knowl-edge sharing and provides incentives. The project managerneeds to establish a climate of trust, where it is safe tomake mistakes (Grant 2006), where sharing knowledge isthe norm and helping others is promoted (Adenfelt &Lagerström, 2006). This nurturing and safe climate isessential for implementing even simple activities such ascollecting accurate and meaningful lessons learned. It is acritical part of project-based learning. One interview par-ticipant noted “it is a rarity that any organization learnsanything from projects themselves.”

Creating a learning climate can be more difficult in aproject team environment than in a permanent organiza-tional unit (Thomas & Tjader, 2000). Project-basedknowledge sharing is limited by pressures of time, con-flicts of interest, and hierarchical project structures. Thegoal is to create a project climate of learning together, onethat cuts across the individual norms and practices thataccompany project members from different organiza-tions and disciplines. Research suggests that the projectteam should develop processes to share, generate, andintegrate knowledge and that from these processes, theteam should create a body of collective project learning(Fong, 2003).

To build a climate for learning, project managers canimplement five practices suggested by the respondents:

• Engage the team when building the risk register. Thisearly knowledge-sharing technique can set the tonefor a project that recognizes risks and does not seek toignore them.

• Communicate that mistakes are a natural part of theteam’s growth and learning.

• Reward behavior that supports a learning climate, notjust behavior that results in the right answer.

• Practice using desired team behaviors (e.g., knowl-edge sharing) on minor issues. This provides teammembers with the opportunity to refine creative andsupportive behaviors before a big issue emerges.

• Speak the truth. Don’t duck the issues that team mem-bers are worried about.

One interview participant explained an approach toimplementing this concept: “Instead of rewarding the cre-ator of a cool and smart design, reward the person whospent a couple of hours; even though he didn’t know thearea, helping someone else solve a problem. That is reward-ing the desired behavior. That’s pretty powerful.”

Establish and Maintain Knowledge Levels

To establish the knowledge levels needed, the project man-ager will, at the beginning of the project, identify the skillsand experience that are needed on the team or available tothe team. Then, using skill profiles or previous experience,people are selected for the team. Many project managers willselect individuals they have successfully worked with in thepast to ensure a cultural fit. Some will select individualswith extensive networks as a means to access internal andexternal knowledge.

Once the team is selected, the project manager will needto ensure that a basic level of process and domain knowl-edge is available to all the team. This can be done individu-ally or collectively, through training courses or workshops orjoint tasks, but it should be done systematically. Activitiessuch as methodology workshops, estimation exercises, talkson the history and strategies of the organization, will helpteam members better understand the larger context andprocess of the project.

When team members leave the project, the knowledgethat they brought to the team and the knowledge they accu-mulated while working on the project is potentially lost tothe team. To prevent such loss, the project manager mustanticipate and fill knowledge gaps.

From interviews and research, several practices weresuggested that can prevent or mitigate the knowledge loss:

• Back up or duplicate key roles: Explicit role shadow-ing—or understudying—helps junior people acquiresenior-level skills.

• Use Swiss-army-type team members. These individu-als can perform a multitude of roles throughout theproject’s lifecycle.

• Use a core team that stays together throughout the project.


1414

• Use retention bonuses and/or opportunities for learn-ing to entice team members to stay on the project.

• Delegate work and responsibility downward to diffuseknowledge across the team: This prevents a knowl-edge bottleneck.

• Develop formal introduction procedures for newentrees (Eskerod & Blichfeldt, 2005).

One interview participant had this to say: “For leader-ship positions where key decisions have to be made, have abackup for the team member. The backups have to be qual-ified enough to be able to make decisions … should thatperson not be around.”

Another added: “Ideally, the majority of people who weused through the phases were the people who were in theprevious phase, and basically we would keep key core teamplayers in pretty much every stream through design, build,test, and implement.”

Create Channels for Knowledge Flow

The project manager needs to establish channels, whichenable sharing of knowledge to benefit the project (Grant,2006). Channels can be created that serve varied purposes—for example, channels to bring knowledge to the team fromoutside, channels to create communities of practice withindisciplinary groups, channels between disciplines andbetween organizational units, channels between phases ofthe project and between levels of authority. Each project willhave its own unique challenges that require differential needfor channels.

Nonaka and Konno (1998), by advancing the conceptof “Ba,” the shared context in which knowledge is shared,created, and utilized, raised attention about the need formanagers to create environments in which knowledge canflow. Communities of practice can be established within aproject to disseminate knowledge, particularly knowledgethat is complex or tacit (Ruuska & Vartiainen, 2005;Williams, 2006). Attention also needs to be paid to thestructure of the project since some structures (e.g., decen-tralized, informal) act as enablers to knowledge creationand sharing (Adenfelt & Lagerström, 2006).

There are many ways and places to share knowledge:websites, repositories, team meetings, industry associations,brainstorming sessions, and visiting experts are examples.The important issue for the project manager is to deliberate-ly and systematically create channels that are interactive,easy to activate, and effective.

Our informants suggested establishing such channelsearly in the project. Used in conjunction with other princi-ples of knowledge management, the project team canbecome self-organizing and effective in finding, creating,and transferring the knowledge required to accomplishproject goals.

In general, many of our project manager informantsused informal socialization among team members as a wayto enhance team knowledge sharing (Fernie, Green, Weller,& Newcombe, 2003). Another broad tactic was to increase

the frequency but shorten the duration of team meetingswhen the project is under a lot of pressure. Some specificpractice suggestions included the following:

• Co-locate team members• Design open office spaces so team members are

encouraged to communicate• Establish lunch-and-learn sessions• Encourage informal gatherings• Develop a team newsletter containing both project

and social information• Conduct daily, 15-minute huddle sessions, where key

team members meet during crucial project periods toshare knowledge on key issues and suggest solutions

• Assign project knowledge connectors: These individu-als know where the expert knowledge resides and canpoint the team in the right direction when a knowl-edge need arises.

One interview participant explained that “at start-uptime, I establish the lifelines that we are going to need. Theseare people I know out there, that I could bring in if we getinto trouble.”

Develop Team Memory

One of the most important resources a team has is the con-tent of its collective memory. Some of this shared memorytakes the form of stories and experiences; some involvesmore explicit lists and lessons. Creating and updatingshared memory is an important team activity at the begin-ning, during and at the end of the project (Kotnour,1999;Rus & Lindvahl, 2002; Schindler & Eppler, 2003).

When starting a project, the project team should gath-er to discuss the lessons they have learned from similarprojects. Everyone will learn something different duringthese sessions, but the result should be a shared under-standing of how (and why) the current project will bedesigned and managed.

As key decisions are made, team members need tounderstand the rationale and to be kept informed aboutprogress and about outstanding problems, both in the proj-ect process and the project domain. Updates to the projectdesign need to be documented so that follow-on steps andphases have the benefit of the rationale. Changes in key con-straints, goals, and assumptions need to be shared by allstakeholders who might be affected.

• After delivering the project, project teams capture thelessons learned for three purposes:

• To give individuals the opportunity to provide input, lis-ten to others, and through this process to increase theirlearning and their competency as project members

• To create a shared story about the project, enablinglearning to be aggregated into a whole, providingsome closure and organizational learning for the teamand the organization

• To pass these lessons learned to other project teamsto help them meet their goals more efficiently and effectively.


1515

Schindler and Eppler (2003) provided some valuablesuggestions on how to improve the process for gathering les-sons learned. One interesting suggestion was to perform thelessons learned gathering graphically, (e.g., collecting andstructuring the project experiences along a timeline as aprocess map with mistakes, successes, and insights) and pro-vide workshop documentation in a poster format visible forall staff involved. They also suggested that any debriefingactivity strive to gain a commitment in the sense of actionconsequences from the gathered insights.

The interview participants use the following practices tosupport the building of team memory:

• Capture lessons learned throughout the course of theproject—don’t wait until the end. This practice will cap-ture the knowledge and experience of those team mem-bers who will leave the team before the project ends.

• By widely and openly discussing lessons learned, proj-ect managers can reinforce their commitment toknowledge sharing (i.e., the climate for learning) andcapture more reliable knowledge than they could viapost-project meetings.

• Use journals to record decisions made and the rea-sons behind these decisions. This technique creates adocument or set of documents that helps the projectmanager brief new team members and governancemembers. Ensure that journals are widely availableand current.

• Assign a project knowledge integrator, someone whoformally manages and integrates project knowledge.This is a more senior role than the project adminis-trator, but could be used as a learning vehicle for aperson in this role.

One interview participant said that “[I create] a recordover the lifecycle of the project. I call it a State of the Nationdocument and it gives kind of a historical record of the proj-ect. I use it also as an instrument for bringing new people onboard. I’ve encouraged teams and clients to do it morebecause sponsors change, and everybody forgets who madewhat decision.”

Use the Risk Register

Although senior project managers are familiar with manyof the knowledge-based risks identified in this research,they lack an established methodology for managing ormitigating them. One suggestion is to explicitly place themost important knowledge risks within the risk register.This action brings them to the attention of the projectsponsor or steering committee, and will help the projectmanager put in place plans and resources to monitor andcontrol them. This suggestion to use the risk register as afocal point is based on information obtained from theinterview participants, several of whom have put high pri-ority risks in the register and obtained the money neededto mitigate this risk.

One interview participant explained that: “With thevendor, we were very aware of this knowledge management

risk. So what we paid the vendor to have an understudy. Thatwas done within the risk management part of the project.”

Knowledge-based risks are different than those nor-mally found in a risk register because they focus on internalteam resources, not external events. We suggest, however,that protecting key project team members and their accu-mulated project knowledge is just as important as protectingthe project against exogenous shocks. Our general recom-mendation is that project managers look for ways to includeknowledge risks in the project risk register and into the proj-ect plan to establish contingencies and secure resources.

Our participants identified several practices around theuse of a risk register for knowledge-based risks:

Integrate lessons learned from earlier projects and usethis information to build the risk register. Project managerscan accomplish this from within the team and by using out-siders who have completed similar projects.

• Use all key members of the team to compile the risk regis-ter and to continuously update it throughout the project.

• Anticipate the departures of key team members andput resources in place for seconding and training suc-cessors.

• Identify key knowledge gaps in the team and putresources in place for external assistance. The sameprinciple can be used for key knowledge transferrequirements.

• If the team is not co-located, the risk is that they willnot create shared knowledge around goals, con-straints, etc. Build events and processes into the planto mitigate this risk.

Conclusion

This research has created a framework to synthesize the keyknowledge and learning issues in IT projects. A definition,knowledge typology and an integrated model comprised of10 knowledge-based risks common to IT projects was pro-posed. These risks are grounded in the literature and sup-ported by feedback from audiences and by interviews with15 senior project managers. From this data, five knowledgeprinciples and associated practices have been created.

This research is in its early stages and many evolvingtrends (e.g., virtual teams, outsourced teams) have not beenexplicitly considered. Further elaboration and validation isneeded. However, it is our hope to capture in this frameworkthe diversity and strength of the work that is being done ina wide variety of disciplines and to provide an overview forresearchers and practitioners. By identifying and cataloguingthe knowledge-based risks that project managers routinelyface, we aim to empower them to better manage these risksand thereby to improve IT project performance and organi-zational project competence.

Acknowledgments

Thanks to the Social Sciences and Humanities ResearchCouncil of Canada for their support of this research. Thanksalso to Ms. Siew Yong Wee, whose able research assistance isgreatly appreciated.


1616

References

Adenfelt, M., & Lagerström, K. (2006). Enabling knowl-edge creation and sharing in transnational projects.International Journal of Project Management, 24(3), 191–198.

Alavi, M., & Leidner, D. E. (1999). Knowledge managementsystems: Issues, challenges and benefits. Communications of theAssociation of Information Systems, 1(7), 7–43.

Alavi,M., & Leidner, D. E. (2001). Knowledge managementand knowledge management systems: Conceptual foundationsand research issues. MIS Quarterly, 25(1), 107–136.

Argyris, C., & Schön, D. (1978). Organizational Learning: Atheory of action perspective. Reading, MA: Addison-Wesley.

Arthur, M., De Fillippi, R., & Jones, C. (2001). Project-based learning as the interplay of career and company non-financial capital. Management Learning, 32(1), 97–117.

Barki, H., Rivard, S., & Talbot, J. (2001). An integrativecontingency model of software project risk management.Journal of Management Information Systems, 17(4), 37–69.

Boh, W. F. (2003). Knowledge sharing mechanisms inproject-based knowledge work: Codification versus personal-ization. The 24th International Conference on InformationSystems. Seattle.

Bredillet, C. N. (2004). Projects: Learning at the edge oforganization. In P. W. Morris & J. K. Pinto (Eds.), The WileyGuide to Managing Projects. Hoboken: John Wiley & Sons.

Brown, J., & Duguid, P. (1991). Organisational learning andcommunities of practice. Organisationl Science (March), 40–57.

Brynolfsson, E., & Hitt, L. (1998). Beyond the productivi-ty paradox. Communications of the ACM, 41(8), 49–55.

Cangelosi, V. E., & Dill, W. R. (1965). Organizationallearning: Observation toward a theory. Administrative ScienceQuarterly, 10, 175–203.

Chan, R., & Rosemann, M. (2001). Managing knowledgein enterprise systems. Journal of Systems and InformationTechnology, 5(2), 37–53.

Cohen, W. M., & Levinthal, D. A. (1990). Absorptivecapacity: A new perspective on learning and innovation.Administrative Science Quarterly, 35, 128–152.

Crowston, K., & Kammerer, E. E. (1998). Coordinationand collective mind in software requirements development.IBM Systems Journal, 37(2), 227–246.

De Fillippi, R. (2001). Introduction: Project based-learn-ing, reflective practices, and learning outcomes. ManagementLearning, 32(1), 5–10.

De Souza, K. C. (2003). Barriers to effective use of knowl-edge management systems in software engineering.Communications of the ACM, 46(1), 99–101.

Eskerod, P., & Blichfeldt, B. S. (2005). Managing teamentrees and withdrawals during the project life cycle.International Journal of Project Management, 23(7), 495–503

Faraj, S., & Sproull, L. (2000). Coordinating expertise insoftware development teams. Management Science, 46(12),1554–1568.

Fernie, S., Green, S. D., Weller, S. J., & Newcombe, R.(2003). Knowledge sharing-context confusion and controver-sy. International Journal of Project Management, 21(3), 177–187.

Fong, P. S. W. (2003). Knowledge creation in multidisci-plinary project teams: An empirical study of the processes andtheir dynamic interrelationships. International Journal of ProjectManagement, 21, 479–486.


Gable, G. G., Scott, J. E., & Davenport, T. D. (1998).Cooperative ERP life-cycle knowledge management. The 9thAustralasian Conference on Information Systems, 227–240.

Garrety, K., Robertson, P. L., & R. Badham. (2004).Integrating communities of practice in technology develop-ment projects. International Journal of Project Management,22(5), 351–358.

Grant, K. P. (2006). Leveraging project team expertise forbetter project solutions. Paper presented at the PMI ResearchConference 2006, Montreal.

Henry, R. M., Kirsch L. J., & Sambamurthy, V. (2003). Therole of knowledge in information technology project governance.The 24th International Conference on Information Systems. Seattle.

Heller, M. (2000, December). The ROI of IT. CIO Executive Research Center. Retrieved fromhttp://www.cio.com/research/executive/edit/value.html

Huang, J. C., & Newell, S. (2003). Knowledge integrationprocesses and dynamics within the context of cross-functionalprojects. International Journal of Project Management, 21(3),167–176.

Huber, G. P. (1991). Organizational learning: the con-tributing processes and literatures. Organization Science, 3,88–115.

Keegan, A., & Turner, R. (2001). Quantity versus quality inproject-based learning practices. Management Learning, 32(1),77–98.

Kotnour, T. (1999). A learning framework for projectmanagement. Project Management Journal, 30(2), 32–38.

Leseure, M. J., & Brookes, N. J. (2004). Knowledge man-agement benchmarks for project management. Journal ofKnowledge Management, 8(1), 103–116.

Lundin, R. A., & Soderholm, A. (1995). A Theory of thetemporary organization. Scandinavian Journal of Management,11(4), 437–455.

March, J. G., & Olsen, J. P. (1975). The uncertainty of thepast: organizational learning under ambiguity. EuropeanJournal of Political Research, 3, 147–171.

Majchrzak, A., & Beath, C. (2003). A framework for study-ing participation and mutual understanding in software proj-ects. Working paper.

Meehan, B., & Richardson, I. (2002). Identification ofsoftware process knowledge management. Software ProcessImprovement and Practice, 7, 47–55.

Middleton, C. J. (1967). How to set up a project organi-zation. Harvard Business Review, March-April, 73–82.

Nelson, K. M., & Cooprider, J. G. (1996). The contribu-tion of shared knowledge to I/S group performance. MISQuarterly, 20(4), 409–429.

Nonaka, I. (1991). The knowledge-creating company.Harvard Business Review, 69 (November–December), 96–104.

Nonoka, I., & Konno, N. (1998). The concept of “Ba”:Building a foundation for knowledge creation. CaliforniaManagement Review, 40(3), spring, 40–54.

Packendorff, J. (1995). Inquiring into the temporaryorganization: New directions for project managementresearch. Scandinavian Journal of Management, 11(4), 319–333.

Parker, S. K., & Skitmore, M. (2005). Project managementturnover: causes and effects on project performance.International Journal of Project Management, 23(3), 205–214.

Pedler, M., Burgoyne, J., and Boydell, T. (1991). The learn-

JU N E 2007 PR O J E C T MA N A G E M E N T JO U R N A L 17

ing company. London: McGraw-Hill.Polanyi, M. (1958). Personal knowledge. Chicago:

University of Chicago PressPolanyi, M. (1966). The tacit dimension. London:

Routledge and Kegan Paul.Postrel, S. (2002). Islands of shared knowledge:

Specialization and mutual understanding in problem-solvingteams. Organization Science, 13(3), 303–320.

Reich, B. H. (2004). Knowledge “traps” in IT projects.Project Management World Today, retrieved from http://pmfo-rum.org/library/papers/2004/1112papers.htm#02

Reich, B. H. (2006). Managing knowledge within IT proj-ects: A review of current literature and directions for futureresearch. Working Paper.

Reich, B. H., & Wee, S.Y. (2006). Searching for knowledge inthe PMBOK® guide. Project Management Journal, 37(2), 11–25.

Rosemann, M., & Chan, R. (2000). Structuring and mod-eling knowledge in the context of ERP. Proceedings of the PACIS(Pacific Asia Conference on Information Systems) 2000Conference, Hong Kong, 623–640.

Rus, I., Lindvall M., & Sinha, S. (2001). A state of the artreport: Knowledge management in software engineering. Dataand Analysis Center for Software, ITT Industries.

Ruuska, I., & Vartiainen, M. (2005). Characteristics ofknowledge sharing communities in project organizations.International Journal of Project Management, 23(5), 374–379.

Rus, I., & Lindvall, M. (2002). Knowledge management insoftware engineering. IEEE Software, May-June 26–38.

Sauer, C., Gemino, A., & Reich, B. H. (2006). Managingprojects for success: The impact of size and volatility on ITproject performance. Working Paper.

Sauer, C., & Reich, B. H. (2006). From process to value—The changing face of IT project management. Working Paper.

Schindler, M., & Eppler, M. J. (2003). Harvesting projectknowledge: Review of project learning methods and success fac-tors. International Journal of Project Management, 21(3), 219–228.

Senge, P. M. (1990). The fifth discipline: The art and practiceof the learning organization. New York: Doubleday Currency.

Sense, A. (2003). Learning generators, project teamsreconceptualized. Project Management Journal, 34(3), 4–12.

Söderlund, J. (2002). Managing complex developmentprojects: Arenas, knowledge processes and time.” R and DManagement, 32(5), 419–430.

Standish Group. (2003, March 25). West Yarmouth, MAhttp://www.standishgroup.com/press/article.php?id=2

Stein, E. W., & Vandenbosch, B. (1996). Organizationallearning during advanced system development: opportunitiesand obstacles. Journal of Management Information Systems,13(3), 115–136.

Thomas, J. L., & Tjader, J. (2000). On learning and con-trol—Competing paradigms or co-existing requirements formanaging projects in ambiguous situations. Paper presented atthe 4th Biannual Conference, International Research Network onOrganizing by Projects. Australia.

Tiwana, A. (2003). Knowledge partitioning in outsourcedsoftware development: A field study. The 34th InternationalConference on Information Systems, Seattle, 259–270.

Tiwana, A., Bharadwaj, A., & Sambamurthy, V. (2003).The antecedents of IS development capability: A knowledgeintegration perspective. Proceedings of the 24th InternationalConference on Information Systems, Seattle, pp. 246–258.

Turner, R. J., & Muller, R. (2003). On the nature of theproject as a temporary organization. International Journal ofProject Management, 21(1), 1–8.

Volkoff, O. (2001). A grounded process model of enter-prise system implementation., Unpublished PhD Dissertation,University of Western Ontario.

Von Stamm, B. (2005). Exploiting the KM-innovationconnection: Managing knowledge through informal networks.Knowledge Management Review, 8(4), 28–31.

Walz, D. B., Elam, J. J., & Curtis, B. (1993). Inside a soft-ware design team: Knowledge acquisition, sharing and inte-gration. Communications of the ACM, 36(10), 63–77.

Walsh, J. P., & Ungson, G. (1991). Organizational memo-ry. Academy of Management Review, 16, 57–91.

Wegner D. M. (1987). Transactive memory: A contempo-rary analysis of the group mind. In B. Mullen & G. R. Goethals(Eds.), Theories of group behavior (pp. 185–208). New York:Springer-Verlag.

Wenger, E. (1998). Communities of practice: Learning, mean-ing, and identity. New York: Cambridge University Press.

Weick, K. E. (1991). The nontraditional quality of organi-zational learning. Organization Science, 2(1), 116–124.

Weick, K. E., & Roberts, K. H. (1993). Collective mind inorganizations: Heedful interrelating on flight decks.Administrative Science Quarterly, 38(2), 357–381.

Williams, T. (2004). Identifying the hard lessons fromprojects—easily. International Journal of Project Management,22(4), 273–279.

Williams, T. (2006). How do organizations learn fromprojects? Presented at the PMI Research Conference 2006,Montreal.

Yoo, Y., & Kanawattanachai, P. (2001). Developments oftransactive memory systems and collective mind in virtualteams. The International Journal of Organizational Analysis, 9(2),187–208.

Zack, M. H. (1999). Managing codified knowledge. SloanManagement Review, 40(4), Summer, 45–58.

BLAIZE HORNER REICH, PhD, is a professor at Simon Fraser University, Vancouver, Canada and a visiting associate at

Templeton College, Oxford University. Before pursuing her PhD, Dr. Reich worked for 15 years in the IT industry as a

practitioner and consultant, focusing on the financial services and utilities sectors. She is currently an editorial board

member for several academic journals and a corporate director of two companies. Dr. Reich’s research has been published

in a wide range of journals, including MIS Quarterly, Information Systems Research, Journal for Management InformationSystems and the Project Management Journal. Dr. Reich’s current project-related research focuses on (1) modelling risk

and IT project performance, (2) the evolving role of IT project managers, and (3) the use of knowledge management

concepts in IT project governance.

JU N E 2007 PR O J E C T MA N A G E M E N T JO U R N A L18

PROJECT MANAGEMENT PROCESSESACROSS BORDERS: A COMPARISONOF EU-US CORPORATE SUBSIDIARYPROJECT ACTIVITIES

This paper examines how project man-

agement activities vary in the corporate

subsidiary of a multinational enterprise.

The results of this case study indicate a

pattern of direct, more simplistic activi-

ties in European operations, while a

more complex pattern of project manage-

ment activities operates in the North

American subsidiary. Implications for

this firm’s efforts to transfer processes

between divisions and to become “one

global company” are discussed along

with directions for future research.

Keywords: project management;

globalization; Europe; U.S.

©2007 by the Project Management Institute

Vol. 38, No. 2, 18-29, ISSN 8756-9728/03

A new world order is replacing generations-old patterns of power andprivilege…Twenty-first century business is in the midst of a social and economic revo-lution, shifting from rigid to permeable structures and processes and creating somethingnew: the boundaryless organization (Ashkenas, Ulrich, Jick, & Herr, 2002, p. 1).

T o combat the pervasive effects of seemingly impermeable architectural barri-ers separating stakeholders, many companies are actively exploring ways toorganize “seamlessly” (Kanter, 1995). Although much of the project man-

agement literature in the past has focused on only a few industries such as thecomputer and construction industries, only recently has the research broadened toinclude additional management areas, such as knowledge and strategy (Urli &Urli, 2000). Project management may be a solution to the problems frequentlyencountered by firms that, in attempting to become boundaryless, have no work-ing maps of the global business territory. This is particularly relevant for projectmanagement knowledge transfer of processes, especially in cross-border opera-tions in general, and in a cross-border acquisition, as is the case here.

According to Badaracco (1991), those firms that are the most competitive areusing knowledge and managing it so that they can compete in an era of continu-ous product improvement and rapid new product introduction. The global mar-ketplace now demands increased knowledge that is readily and quickly assessablewith shortened product life cycles, increased product complexity, technologicaladvances, and low inflation limiting the ability of firms to increase prices(Disterer, 2002; Judgev & Thomas, 2002; Pinto, 2002).

Knowledge migrates quickly, as IBM discovered once it made public theworkings of its personal computer (Badaracco, 1991). The firm watched its shareof the PC market steadily erode as clone manufacturers geared up to producewhat quickly became a commodity. Given the tremendous costs involved indeveloping and refining new technologies and speeding new ideas to market, thesimultaneous “compete and collaborate” model that is well-practiced in Japannow also is being employed by many U.S. corporations. Thus, more and morefirms are attempting to revise their concept of traditional boundaries and to thinkin terms of a more fluid business environment and organization through a vari-ety of methods, including collaborations with competitors (Hirschhorn &Gilmore, 1992; Sandelands, 1994), as well as other more common transfers ofknowledge across divisions of the same firm. Firms that are able to work with

PHILIPPE BYOSIERE, Doshisha University, JapanDENISE J. LUETHGE, Drake University, Iowa

ABSTRACT


these projects, perhaps the most usefuldefinition, and one that has beenbased upon years of project manage-ment research and study, is that pro-posed by the Project ManagementInstitute. According to the PMI’s AGuide to the Project Management Body ofKnowledge, or the PMBOK® Guide,“Project management is the applica-tion of knowledge, skills, tools, andtechniques to a broad range of activi-ties in order to meet the requirementsof a particular project. Project manage-ment is accomplished through theapplication and integration of theproject management processes of initi-ation, planning, executing, monitoringand controlling, and closing” (PMI,2004, p. 8). Cleland and King (1983),Ford and Randolph (1991), Disterer(2002), and PMI (2004) definitionsserve as useful background for the pur-poses of this paper in that they encom-pass the many varieties of themanagement of projects that haveappeared in the literature as well as inthe case study firm.

This research is exploratory innature and examines project manage-ment across boundaries within anMNC, as the top management teamexamines those criteria that distinguishbetween successful project teams andunsuccessful teams. As such, it is usefulto look at the literature as it relates toboth project lifecycles as well as criticalsuccess factors (CSF) in project man-agement processes.

A four-stage lifecycle model dis-cussed by Adams and Barndt (1983)and further developed by Slevin andPinto (1987) and Pinto and Prescott(1990) serves as a useful categorizationscheme for this exercise. This particularmodel was discussed at length inJugdev and Müller’s (2005) survey ofproject success and how it has beendefined over the past 40 years. Theproject lifecycle describes the initia-tion, operations, and retirement of aproject (PMI, 2004), for which thereare a number of conceptualizations(Jugdev & Müller, 2005). In brief, thefour stages are: (1) conceptualization,defined as the stage where top man-agement acknowledges the need topursue a strategic issue and identifies

these more fluid structures, such asDell, are less constrained to the “oldways of doing business” and are ableto take advantage of market changes.

One method in particular that hasoften been viewed as a promising toolfor knowledge creation and “bound-ary-blurring” is project management. Ifdesigned and implemented properlywith mutually agreed-upon, well-defined goals, project management notonly can help firms neutralize the fre-quent intradepartmental rivalriesbetween functional “silos,” but it alsocan help forge linkages between prod-uct group or divisional “fiefdoms” andeven help bridge strategic alliance orjoint venture partners, enhancing thecreation and dissemination of knowl-edge and the exchange of other criticalresources in the process (Al-Ghamdi,1998; Cheng, Love, & Irani, 2001;Goodwin & Elliott, 1995; Gorman,2002; Skok & Legge, 2002; Wessel,1993). As a result, project manage-ment can serve as a useful tool forimplementing corporate strategies,especially as it aids in the systematictransfer of knowledge from one projectto another and one stakeholder toanother (Archibald, 1987; Beck, 1983;Blessing, Goerk, & Bach, 2001;Disterer, 2002; Gorman, 2002; Judgev& Thomas, 2002; Snider & Nissen,2003; Weiser & Morrison, 1998).

This paper examines the utility ofproject management within a large U.S.multinational corporation (MNC) as astrategic tool being employed to dissolverigid boundaries and enhance the flowof knowledge to assist the firm in itsendeavors to become the global leaderin its business. More specifically, it pres-ents a comparative analysis of the rela-tionships among a set of projectmanagement activities across the projectlifecycle for the North American andEuropean operations of the MNC. Tothis end, this paper reviews the literatureon project management to arrive at anappropriate definition of the concept asbackground for this case study.“UltraCleen“ (a pseudonym for theactual company) is introduced as a casestudy. This is a U.S.-based Fortune 100MNC that recently acquired a large divi-sion of a European firm as part of its

globalization strategy. Next, the roleproject management is playing to assistUltraCleen in its efforts to share knowl-edge, break down bureaucratic bound-aries, and become “one globalcompany” is discussed. The results ofthe comparative analyses of global proj-ect management activities within theNorth American and European opera-tions is presented, and the implicationsfor UltraCleen’s efforts to fulfill its visionof becoming the global leader in its linesof business through the next decade isdiscussed. The paper concludes withdirections for future research.

A Literature-Based Definition of

Project Management

A cursory review of the extensive litera-ture on the topic reveals a vast numberof project management definitions.The concept generally entails bringingpeople together within similar or dif-ferent departments, divisions, or evencompanies to work on a specific taskor project, upon completion of whichthe team generally disbands. Ford andRandolph (1992, p. 268) concludedthat project management “has come tomean a variety of things” and like thematrix organization, “this variety ofuses and forms makes it difficult tofind consensus on a concise and pre-cise definition.” Cleland and King(1983, p. 187) defined project man-agement as a “combination of humanand nonhuman resources pulledtogether in a ‘temporary’ organizationto achieve a specific purpose.” Fordand Randolph (1992, p. 272) pro-posed their own definition that blendsaspects of a matrix organization withproject management. In this respect,the authors define project manage-ment as “cross-functional organiza-tional overlays that create multiplelines of authority and that place peo-ple in teams to work on tasks for finiteperiods of time.”

Disterer (2002, p. 513) definesprojects as “temporary organizationswith specific objectives, detailed tasks,and restricted time and budget. Whena project is finished, normally there isno institution or corpus left whereexisting knowledge can be accessed.”When looking at the management of


preliminary goals and alternatives aswell as organizational capability insuccessfully carrying out the project,(2) planning, defined as the formaliza-tion of concrete plans to achieve thepreliminary goals and the allocation ofspecific tasks and resources, (3) execu-tion, defined as conducting the worknecessary to transform plans intoresults, and (4) termination, defined ascompletion of initial objectives andthe reassignment of resources. Theteam merged a fifth stage, called evalu-ation, with termination, to measureproject outcomes against project objec-tives. Although different classificationsexist (see, for example, Turner, 1999),all are quite similar, and we will referto the classification in Slevin and Pinto(1987) for the remainder of this paper.

In addition to the stages that occurduring the life of a project, numerousauthors have examined factors that arecritical to the success of a project.Fortune and White (2006) reviewed 63publications that focus on CSFs. Theyidentified support from senior man-agement, clear realistic objectives, astrong/detailed plan kept up to date,good communications/feedback, anduser/client involvement as the top fivesuccess factors across all of these stud-ies. Fortune and White (2006) com-pleted an extensive review, which alltheir findings will not be repeatedhere, but the reader is directed to thisvery thorough review of the criticalsuccess factors literature. However,their findings are consistent with addi-tional research by Taylor and Morris(2005), Drummond and Hodgson(2003), Guttman and Longman(2006), Smith (2003), Rad (2002),and Kanter and Walsh (2004), whichshow that many of these success fac-tors relate not so much to the deliver-ables but to the process of projectmanagement.

UltraCleen and Its Path to Globalization

UltraCleen is a century-old Fortune100 firm with a national reputation forits commercial and retail home fur-nishings products. Until recently, a rel-atively small percentage of its grosssales were from overseas operations. Itsmanufacturing operations were prima-

rily located in the U.S., Canada,Mexico, and Brazil. In the mid-1980s,with the onslaught of globalization,UltraCleen’s newly installed CEOcame to the conclusion that unless thefirm radically changed, its independ-ence and survivability was uncertain.Under the two previous CEOs, theportfolio of businesses had been man-aged as a “cash cow.” With amplefunds in its corporate coffers,UltraCleen was easy prey as a leveragedbuy-out or takeover target.

Taking stock of the company’sstrengths and weaknesses, and follow-ing an evaluation of opportunities andthreats in the external environment,UltraCleen’s CEO crafted a boldvision: global leadership in the 21stcentury. The path to this lofty goalcalled for selective domestic and inter-national acquisitions and more effi-ciently and effectively managedinternal growth. The CEO then sharedhis proposed strategy with members ofthe board of directors and senior man-agement, who, after considerable dis-cussion, largely concurred. Thus werespawned search initiatives that ulti-mately led to the acquisition of a largedivision of a European conglomerate.The deal, completed in incrementalownership stages over the course ofseveral years, nearly doubled the sizeof the company and overnight provid-ed UltraCleen with a strong marketposition within Europe. As a result,more than 40% of UltraCleen’s rev-enues were now generated from over-seas operations.

With the acquisition finalized,UltraCleen’s CEO faced the dauntingtask of forging both organizations into“one global company,” a critical firststep toward fulfilling his vision. Oneof his first initiatives was to assembleall of the vice presidents of the organi-zation for a week-long event during themid-1990s, which had dual purposes:socialization of the top managementteam (TMT), whereby North Americanand European vice presidents couldbegin to get to know one another, andthe identification of the most criticalissues facing the organization in itsdrive to become the global leader in itsindustry. The exercise resulted in the

generation of more than 20 criticalissues, which were reduced to 15through extensive discussion and cate-gorization. Then, global cross-func-tional teams led by of nine vicepresidents in the organization wereformed and charged with thoroughlyexamining one of the 15 critical issues,identifying meaningful alternativecourses of managerial action, andreporting their findings to their col-leagues in the next company-widemeeting 12 months hence.

Project Management as a Strategic

Issue at UltraCleen

One of the 15 key priorities deemedcritical to the company’s long-termperformance and fulfillment of itsstrategic goals was global project man-agement. The issue was prioritized as itwas widely believed that well-definedand structured project managementinitiatives carried out on a global levelthat were sponsored by UltraCleen’sCEO and/or the CEOs of the NorthAmerican and European operationscould (1) facilitate the integration ofthe new acquisition, (2) break downtraditional barriers or boundariesbetween organizational units, (3)enhance the transfer of knowledge,and thus (4) contribute to organiza-tional learning to achieve and to sus-tain competitive advantage.

To intensively study the topic, anine-member Global ProjectManagement Team (GPMT) wasformed. The team was comprised of theEuropean and North American vicepresidents for manufacturing and mar-keting, the corporate vice president ofoperations and planning, the European,North American and Corporate vicepresidents for information technology,and four manufacturing division vicepresidents. In keeping with the “oneglobal company” goal, the nationalitiesof the team members were intentional-ly diverse by design: three vice presi-dents were from the U.S., two were fromItaly, and one each was from Canada,Spain, Sweden, and Mexico.

The GPMT embarked on its mis-sion by first establishing global projectmanagement guiding principles tosubsequently assist managers in the


managers within UltraCleen wereasked to rank-order (1) the criteriathey felt “creates” a project, (2) whomakes that determination, and (3) keysuccess factors for project manage-ment. The participants were also askedto list two projects completed over thepast year that they felt were successfuland two projects that were unsuccess-ful and then to rank-order the reasonswhy. The third strategy, benchmarking,involved identification and assessmentof global project management prac-tices within other MNCs located inNorth America, Japan, and Europe inorder to ascertain key learnings and todisseminate “best practices.”

The focus of this paper is on theresults of the analysis conducted toaddress the first strategy, project assess-ment. This entailed a comprehensivecomparative assessment of processactivities associated with 20 multimil-lion-dollar global projects originatingin the North American or Europeansectors of the business and sponsoredby the CEO and/or his immediatedirect reports.

Methodology

A research team was assembled to pro-vide assistance with instrument devel-opment and data analysis regardingactivities associated with the 20 globalprojects. Drawing from the literature aswell as from its knowledge of projectmanagement practices withinUltraCleen’s North American andEuropean businesses, the researchteam generated a list of 31 items orprocess activities reflecting differentstages of the project lifecycle, of which26 are reported here. The remainingfive items were developed specificallyfor the management team and unrelat-ed to this analysis. A two-step proce-dure was followed for data collection.First, a pre-interview guide was devel-oped that was a structured, open-ended survey instrument comprised ofspecific questions pertaining to theprocess activities reflecting the projectmanagement stages previouslydescribed. A 5-point Likert-type scalewas employed to capture the partici-pants’ assessment of each processactivity comprising the project stages.

design and implementation of global,rather than national, products. Morespecifically, these guiding principleswere intended to aid participatingproject team members in clarifying thedefinition of projects with worldwideimplications and the appropriate rolesof the various stakeholders (end-user,customer, sponsor, project team),highlighting the most effective projectmanagement organization structures,and identifying measures for evaluat-ing project management.

Defining Project Management Within

UltraCleen

The GPMT approached their assign-ment by developing a common defini-tion of project management—that is, adefinition they felt would appropriate-ly capture the variety of initiativesunder way in the company—whichthey supplemented with a glossary ofall major terms connected with theconduct of project management in aglobal context. They defined projectmanagement as “a cross-functionalteam effort requiring significantresources that has a start and an end,and a tangible result to the end userthat is predefined in a contractbetween the sponsor, the customer andthe project team.” Clearly, the GPMT’sdefinition is similar to the literature-based definitions previously present-ed, though it differs in its clarificationof end-results and inclusion of con-tractual agreements between key proj-ect stakeholders.

The project team was defined as thegroup of people who are responsibleand accountable for providing theproject deliverables as specified by thecontract and/or subsequent contractchanges. The end-user was defined asthe downstream person who actuallyuses the product(s) or service(s) deliv-ered from the project. The project spon-sor was defined as a member ofmanagement (director or higher level)who has stated a commitment to theproject and who has the authority toapprove or influence the approval ofmeasures, resources, etc., required tocarry out the project. The project directorwas defined as the individual who wasin charge of coordinating the day-to-

day activities of the overall project. Thecustomer was defined as the personwho requests the project and would beresponsible for the management oroperation of the product(s) or serv-ice(s) delivered from the project. Thisperson is either the end-user, or a rep-resentative of the end-user. The steeringgroup within UltraCleen was defined asa taskforce of senior level executiveswho determined the “fit” of the projectwithin the larger framework of the cor-poration and its strategy. Finally, theGPMT defined the contract as a formalagreement between the customer andthe project team.

The GPMT then identified threeinterdependent strategies that wouldenable them to answer the question ofwhat was effective and successful glob-al project management throughoutUltraCleen. The first strategy, which wewill call project assessment, entailed acomprehensive assessment of majorglobal projects in the North Americanand European operations that wereimplemented during the two-year peri-od in which UltraCleen first acquired amajority share of the European compa-ny until it assumed complete owner-ship. These projects were global interms of their design and objectives,crossed international boundaries, tend-ed to involve managers from differentsites as well as functions, and were ofsuch importance that they were spon-sored by the CEO and/or his immedi-ate direct reports. The methods ofassessment that the GMPT employedwere intensive qualitative interviewswith the project directors assigned tothese global projects. The project direc-tor had to be officially appointed, havefull-time responsibility for the projectand managing the project team, andhave sufficient and recognized authori-ty in the organization.

The second strategy, knowledgeidentification, entailed cataloging andassessing past and current projects,both domestic and international,which were identified either as success-ful or unsuccessful by the entire seniormanagement team to distill key learn-ings. The methodology employed wasthe administration of a structuredquestionnaire in which the top 140


Second, each project director was thenpersonally interviewed by a member ofthe research team to further explore hisor her responses to the survey andrelated project management issues.When changes occurred in projectdirectors over the course of the projectfor whatever reason, interviews wereconducted with both the departingand succeeding project directors priorto the actual completion of the project.Also worth noting is that approximate-ly 25% of the projects identified hadmore than one project director. Thedata for each of these projects wereaggregated, hence the level of analysisis for the project rather than for theproject director.

The objective of this initial exer-cise was to identify similarities and dif-

ferences that emerged in the projectmanagement activities between theNorth American and European proj-ects. Of the 20 global projects targetedby the GPMT, 14 originated in NorthAmerica and 6 in Europe. They includ-ed the development of a new productline, examination of the distribution ofproducts, elimination of redundanciesin positions caused by the acquisition,development of worldwide leadership,and the formulation of a new compen-sation strategy. Table 1 presents select-ed examples of three European andthree North American projects alongwith a concise description of the proj-ect and objectives, identification of theorigin of the project and its sponsor(s),and a summary of key driving and con-straining forces.

Analysis

Formulating and completing 20 globalprojects within a two-year time frame(mid-1990s) in one sense is a consid-erable achievement. Nonetheless, theprojects were still few in number,which places restrictions on the abilityto utilize powerful statistical tech-niques for analyzing the data and togeneralize beyond this sample or thisfirm. The authors’ analytical planinvolved (1) the computation of uni-variate statistics for all variables, i.e.,testing the differences in the means foreach individual variable, (2) the com-putation of Mann-Whitney U tests andMedian tests to examine differences incentral tendencies between theEuropean and North America samplesfor each of the variables (Siegel, 1956),

Table 1: Selected examples of European and North American global projects within UltraCleen


and (3) the calculation of Pearsonproduct moment correlations and thecorresponding development of correl-ograms, geospatial depictions of thecorrelation or association between twodifferent variables, given that these cor-relations yield stronger statisticalpower for smaller sample sizes(Kraemer & Thiemann, 1987).

Results

Table 2 presents the means and medi-ans computed for 26 variables for eachof the 14 North American projects andthe 6 European projects as well asMann-Whitney U test results. Exceptfor one variable, the consideration of

risk during the planning phase of theproject, no significant differences inMann-Whitney U tests were foundbetween the European and NorthAmerican samples, indicating that themeasures of central tendency of the 26variables are not significantly differentfor the North American and Europeanprojects. Although there are few indi-vidual variable differences, there maybe significant differences in relation-ships within the project managementprocesses. To examine these, we look atcorrelograms of the U.S. versusEuropean projects.

To facilitate the interpretation andpresentation of the correlational analy-

ses, correlograms were developed,which illustrate the significant rela-tionships between the activities associ-ated with the European and the NorthAmerican projects based on thePearson correlations. Figure 1 presentsthe overall significant relationships forboth the European and NorthAmerican divisions between the fourstages of project management: concep-tualization, planning, evaluation, andexecution. In Europe, the focus is onplanning and execution. The evalua-tion stage significantly impacts onlythe conceptualization stage, and theplanning stage impacts only the execu-tion stage. This is very different from

Table 2: Mean, median, and Mann-Whitney U test results


the North American situation. In thiscase there is a much more complexprocess and there are many more con-nections. In this case, both the concep-tualization stage and the executionstage are linked to both the evaluationand planning stages. Interestingly, theconceptualization stage is not linkedto the execution stage, as it is in theEuropean case, which will be discussedfurther in a later section.

Figure 2 shows the correlogramfor the European projects in whichproject management activities are cat-egorized by the four lifecycle stagespreviously described. Only correla-tions significant at the .01 level orgreater are displayed, as the authorschose to use a stronger level of signifi-cance due to sample size limitations.As previously discussed, the evalua-tion stage impacts only the conceptu-alization stage, and the executionstage ties back to the conceptualiza-tion stage only for the project directorand team; that is, no third parties areinvolved. Interestingly, no linkagesbetween the evaluation and planningstages or between the evaluation andexecution stages are shown.

As indicated in Figure 2, there isno significant correlation reported byEuropean managers between the con-ceptualization and planning phases,

nor between the execution and evalu-ation phases. Unlike what has beendescribed in the literature, theEuropeans do not seem to think of theproject management process as aseries of sequential steps. Specifically,the success of the project’s sponsorleadership is determined by the needfor project sponsor leadership (.96).In addition, the consensus amongproject teams is strongly related to theneed for peers of sponsor leadership(1.00). Thus, it would seem that thereis high consensus among the projectteam members when there is a highneed for sponsor leadership. It is pos-sible that the project team may feelthat sponsor peer support is indicativeof a project mandate, thus stimulatinggreater support from the team mem-bers as well.

In terms of the execution andplanning stages, European managersbelieve that participation between anumber of stakeholder groups can begreatly hindered by the extent to whichdaily operations inhibit the comple-tion of the project (-.93 and -.93).Furthermore, the execution stage alsois linked with the conceptualizationstage in that participation between theproject director and the team is signif-icantly correlated with the sponsor’sdefinition of project output (.89). In

other words, the more clearly thedeliverables are defined by the projectsponsor (senior management), thehigher the level of participation willbe between the project membersresponsible for getting things done.

Figure 3 illustrates the relation-ships between the North Americanproject activities by lifecycle stage. Inthe United States, the process is muchmore complex, involving many morevariables and relationships. In thiscase, the focus is on the evaluationstage and its relationship with allother aspects of the project manage-ment process, with linkages betweenthe project evaluation, conceptualiza-tion, and planning stages being partic-ularly significant.

There are too many significantrelationships to discuss each of themindividually, and as a result, the fivelinkages between the four phases of theproject are summarized. In the firstlinkage, there are four project leader-ship variables that connect the concep-tualization phase with thetermination/evaluation phase, connect-ing all aspects of leadership (projectdirector, sponsor, and sponsor peers)with the success of the project directoras well as with anticipated team per-formance (A, B, C, and D with G). Thissuggests that up-front agreement

Figure 1: European vs U.S. project management activities


and evaluation phases of the project.In this case, the focus is on communi-cation between the project directorand the team. Participation betweenthe project director and the team issignificantly related to consensusbetween these two, but also to thesuccess of the team and consensusamong the team members (P with I, J,and K). Also, participation among theproject team members is significantlyrelated to consensus among teammembers and the director (S with K).It is evident that open communica-tion with the team and the projectdirector as well as commitment to theproject via participation enhancesconsensus and successful perform-ance of the project team.

Discussion

A review of the correlograms illus-trated in Figures 1, 2, and 3 revealsthat there are few common relation-ships among the 26 project activi-ties in the European and NorthAmerican samples. One commonfinding pertains to the relationshipbetween need for sponsor leader-ship at the conceptualization stageand success of sponsor leadership atthe evaluation stage for each sample

between the project sponsor and proj-ect director is associated with highconsensus levels among project direc-tor, project sponsor, and project teammembers. In addition, agreementbetween the project director and proj-ect sponsor as well as having a well-defined output by the sponsor in theconceptualization phase is related toconsensus between the project directorand the project team (E and F with K).In other words, when the project out-put is well-conceived and the projectdirector “buys in” to the sponsor’sproject conceptualization, there is agreater consensus between the projectdirector and the team.

In the second linkage, the concep-tualization phase is also connected tothe planning phase, with the projectdirector and project sponsor variablesin the conceptualization phase linkedwith those variables addressing com-pany project support at the planninglevel (E with L and M; F with M). Thissuggests that project legitimacyachieved through strong agreementbetween the project director and overtsupport by project sponsor is associat-ed with high levels of consensusamong project team members. Thethird linkage shows the relationship

between the planning and evaluationstages, where the success of projectdirector leadership is significantlyrelated to the planning of the projectdirection by the project sponsor, thus,this relationship stresses the necessityof leadership between planning andevaluation (O with G). In addition,project support variables in the plan-ning stage are significantly linked tothe evaluation stage with consensusamong team members, but consensusbetween the sponsor and the team ishurt by daily operations which inhibitthe project (L with J, M with H). Thus,communication and consensus amongteam members is vital for driving theproject through to completion as wellas continuing support of both manage-ment and the team.

The fourth linkage, connectingthe planning and execution phases,links the planning of the projectdirection by the project sponsor withactual project sponsor involvementwith the team (O with Q and R). Thisis an important relationship that wedo not see in the European case,where communication between theproject sponsor, project director, andthe team is significant. Finally, thefifth linkage is between the execution

Figure 2: European division


(see Table 2 for the exact wording ofthe variables in question). A secondcommon finding is that levels of partic-ipation and consensus tend to beinhibited by daily operations in boththe European and North American sec-tors of UltraCleen’s businesses. Third,neither correlogram reveals a clusterwith significant relationships betweenproject activities across all four stages ofa project’s lifecycle. This would seem tobe an opportunity for each of the divi-sions to examine the potential forknowledge transfer in areas where nolinkages currently exist.

The differences between theEuropean and North American projectactivity relationships are greater innumber than the similarities. As notedearlier, no linkages are shown betweenthe evaluation and planning stages orbetween the evaluation and executionstages in the European sample. In part,this may be the result of the structureand organization of the company’sdivisions in Europe. In this case, thedivisions are country-based andbrand-name based. This means that

many managers tend to communicateonly with their national peers, ratherthan to communicate with othersponsors, peers, or directors fromother countries within Europe oracross the regions. The variables in theconceptualization phase are related tothe evaluation of time, cost, ease ofimplementation, and quality of theproject at completion for theEuropean sample only. There is onlyone planning-stage variable related toother lifecycle variables within theEuropean correlogram, whereas onlyone execution-stage variable is relatedto project lifecycle variables within theNorth American correlogram.

In the case of the United States,the only stage relationship missing isthat between the conceptualizationand execution stages, most likelybecause there is such a strong rela-tionship between the conceptualiza-tion and planning stages, resulting insome redundancy. In addition, thestructure of the company in the U.S.in this case study is product-based.This facilitates communications

across regions because the globalprojects are not product-based, andtherefore the required information tocomplete the project may be sourcedoutside of the local organization, andin many cases, outside of the regionas well, thus fostering knowledgetransfer across borders.

The influence of the sponsor inNorth America seems to be critical tothe success of the project and the rela-tionships among the project stake-holders, especially the relationshipbetween project sponsor and projectdirector. And in contrast with theEuropean sample, evaluation of theproject is primarily expressed throughrelationships reflecting communica-tion among the project director, proj-ect sponsor, and project team for theNorth American sample.

Implications for Multinational

Enterprises

Although this study is not a complex sta-tistical analysis of an examination acrossmultiple companies, it is a humbleattempt to provide an in-depth analysis

Figure 3: North American division


of the inner workings of the manage-ment of projects within two large sub-sidiaries of a multinationalorganization. In this case, we see a rich-er complexity of the process within theU.S. operations, as opposed to the sim-plicity, and perhaps efficiency, of theEuropean operations, which appear tomove in a more lock-step fashion.These results have implications formultinational enterprises (MNEs) withprojects in multiple locations. Theysuggest that there is a discerniblesequence of project management activ-ities that reflect different stages of theproject lifecycle within the firm, andthat learnings in one division can beapplicable to other divisions. However,these activities, and even stages, may bereciprocal. That is, planning or perhapsexecution outcomes may call for thereconceptualization of the project,which may lead to restarting the cycle.New data (e.g., new technologicaladvances, competitive intelligence, etc.)may become available at any stage ofthe cycle and may thus influence activ-ities at that or subsequent stages.

This paper’s findings also suggestthat within this global technology-ori-ented company, a clear definition ofthe roles of stakeholders is important,and in particular, having an upper-level sponsor for a project that crossesnational boundaries is critical to suc-cess. Relatedly, global project manage-ment is more than the sum ofEuropean and North American projectmanagement activities. It is morecomplex, especially given technical,language, and cultural differences.However, global project managementclearly has utility as a strategic toolthat UltraCleen’s CEO can effectivelyuse for dissolving structural barriers,facilitating the integration of theEuropean acquisition, and enablingboth the North American andEuropean operations to share knowl-edge and resources critical to theirmutual success and effectiveness. Thisis especially critical, given thatUltraCleen stresses its focus upontechnology, and many of the projectsinvolved in this study involved thetransfer of knowledge either with thetechnology itself, processes related to

the technology, or the management ofthese projects or processes.

To maximize the potential ofglobal project management’s utilityas a strategic integration and knowl-edge creation/dissemination tool,UltraCleen should examine whythere are fewer similarities than dif-ferences in relationships betweenproject management activitiesbetween the European and NorthAmerican operations. Addressing thisissue may shed light on why, as well,none of the emergent clusters incor-porates variables from all four projectlifecycle stages. A useful startingpoint would be the evaluation of datacollected with regard to the successand failures of projects, both domes-tic and global, which originate eitherin the European or North Americanoperations. The results of the interna-tional project management bench-marking study under way also couldprove useful in clarifying key issuesand identifying areas warrantingattention to enhance the potential ofglobal project management as astrategic tool.

Limitations to the Study

There are several limitations to thisstudy that may affect the generalizabil-ity of our findings. First is samplesize—the small number of globalprojects of this magnitude that hadbeen conducted within UltraCleen atthe time of this study. Second, theinstrument that was developed toreflect different stages of the projectlifecycle did not capture other criticalactivities or issues, such as projectteam composition, team skills pro-files, language barriers, communica-tions channels, and other national orcorporate cultural variables. Althoughthe level of analysis is the project, ide-ally the study would have beenstronger if views from different stake-holders such as team members, end-users, sponsors, etc. had beenrepresented. Relatedly, because theglobal project management team wascomprised of vice presidents from thecompany, there might have been abias in their responses to the ques-tionnaire items that could have affect-

ed results, given that the global projectwas the level of analysis in this study.

Directions for Future Research

Though project management has beenstudied for over three decades, therehas been little attention devoted to (a)projects that are global in scope andare used as strategic tools to dissolveorganizational boundaries and cre-ate/disseminate knowledge, and (b)consideration of multiple stakehold-ers. This study is one of very few thatfocuses on the role and implications ofglobal project management as a toolfor implementing strategies for global-ization with consideration of sponsors,directors, project team members, andclients/end-users. Though the studywas based on a small number of glob-al project management initiativesundertaken within one multinationalfirm, the results suggest that there arecontent and process differencesbetween the North American andEuropean operations of this firm thatmay warrant managerial attention andmay be useful intervention points toenhance project success. This clearlypoints to the need for larger-scale stud-ies, and those studies examining rela-tionships across multiple cultures.

Given the globalization of virtu-ally all industries—services and man-ufacturing alike—there is a call formore theoretical and empiricalresearch on global project manage-ment, especially as it relates to organi-zational structure and strategy. Thereplication of this study, with refine-ments made to instrumentation toaddress some of the deficiencies pre-viously described that would alsoinclude project performance variables(e.g., content and process) would be auseful starting point. Comparativeglobal project management researchwith a sample of international firmsin various industries also wouldmeaningfully contribute to our com-prehension of the design, implemen-tation, and successful completion ofprojects with a global focus. Takingthese steps, in the authors’ view,would contribute further to differenti-ating the map from the territory of theboundaryless firm.


Acknowledgments

The authors would like to thank theInstitute for Technology, Enterprise andCompetitiveness at DoshishaUniversity, Kyoto, Japan, for its gener-ous support of this research. In addi-tion, the authors would like to thanktwo anonymous PMJ reviewers for theirmost helpful comments.

References

Adams, J. R., & Barndt, S. E. (1983).Behavioral implications of the projectlifecycle. In D. I. Cleland & W. R. King(Eds.), Project Management Handbook(pp.222–244). New York: Van NostrandReinhold Co.

Al-Gahmdi, S. (1998). Obstaclesto successful implementation of strate-gic decision: The British experience.European Business Review, 98(6),322–327.

Archibald, R. D. (1987).Implementing business strategiesthrough projects. In W. R. King & D. I.Cleland (Eds.), Strategic Planning andManagement Handbook(pp. 499–507).New York: Van Nostrand Reinhold Co.

Ashkenas, R., Ulrich, D., Jick, T., &Herr, S. (2002). The boundaryless organi-zation: Breaking the chains of organiza-tional structure. San Francisco:Jossey-Bass, Inc.

Badaracco, J. L. (1991). The knowl-edge link: How firms compete throughstrategic alliances. Boston, MA: HarvardBusiness School Press.

Beck, D. R. (1983). Implementingtop management plans through projectmanagement. In D. I. Cleland & W. R.King (Eds.), Project ManagementHandbook(pp. 166–184). New York:Van Nostrand Reinhold Co.

Blessing, D., Goerk, M., & Bach, V.(2001). Management of customer andproject knowledge: Solutions and expe-rience at SAP. Knowledge and ProcessManagement, 8(2), 75–90.

Cheng, E., Li, H., Love, P., & Irani,Z. (2001). Network communication inthe construction industry. CorporateCommunications, 6(2), 61–70.

Cleland, D. I., & King, W. R.(1983). Project management handbook(3rd ed.). New York: Van NostrandReinhold.

Disterer, G. (2002). Managementof project knowledge and experiences.Journal of Knowledge Management, 6(5),512–520.

Drummond, H., & Hodgson, J.(2003). The chimpanzees’ tea party: Anew metaphor for project managers.Journal of Information Technology,28(September), 151–158.

Ford, R. C., & Randolph, W. A.(1992). Cross-functional structures: Areview and integration of matrix organ-ization and project management.Journal of Management, 18(2),267–294.

Fortune, J., & White, D. (2006).Framing of project critical success fac-tors by a systems model. InternationalJournal of Project Management, 24,53–65.

Goodwin, J., & Elliott, C. (1995).Exporting strategies: Developing astrategic framework. SAM AdvancedManagement Journal, 60(1), Winter,21–28.

Gorman, M. E. (2002). Types ofknowledge and their roles in technolo-gy transfer. Journal of Technology Transfer,27(3), 219–231.

Guttman, H., & Longman, A.(2006). Project teams: How good arethey? Quality Progress, 39(2), 59–65.

Hirschhorn, L., & Gilmore, T.(1992). The new boundaries of the“boundaryless” company. HarvardBusiness Review, May-June, 104–115.

Jugdev, K., & Müller, R. (2005). Aretrospective look at our evolvingunderstanding of project success.Project Management Journal, 36(4)19–31.

Judgev, K., & Thomas, J. (2002).Project management maturity models:The silver bullets of competitive advan-tage? Project Management Journal, 33(4),4–14.

Kanter, R. (1995). World-class:Thriving locally in the global economy.New York: Simon & Schuster.

Kanter, J., & Walsh, J. (2004).Toward more successful project manage-ment. Information Systems Management,21(2), 16–21.

Kraemer, H.C., & Thiemann, S.(1987). How many subjects? StatisticalPower Analysis in Research. Beverly Hills,CA: Sage Publications, Inc.

Pinto, J. (2002). Project manage-ment 2002. Research TechnologyManagement, 45(2), 22–37.

Pinto, J., & Prescott, J. (1990).Planning and tactical factors in projectimplementation success. The Journal ofManagement Studies, 27(3), 305–328.

Project Management Institute(PMI). (2004). A guide to the projectmanagement body of knowledge, 3rd edi-tion, Newtown Square, PA: ProjectManagement Institute, Inc.

Rad, P. (2002). A model to quanti-fy the success of projects. AACEInternational Transactions, CS05.1-05.4.

Sandelands, E. (1994). All talkand no action? Perish the thought.Management Decision, 32(5), 10–11.

Siegel, S. (1956). Nonparametric sta-tistics for the behavioral sciences. Tokyo:McGraw-Hill Kogakusha, Ltd.

Skok, W., & Legge, M. (2002).Evaluating enterprise resource plan-ning (ERP) systems using an interpre-tive approach. Knowledge and ProcessManagement, 9(2), 72–82.

Snider, K., & Nissen M. (2003).Beyond the body of knowledge: Aknowledge-flow approach to projectmanagement theory and practice.Project Management Journal, 34(2),4–12.

Slevin, D., & Pinto, J. (1987).Balancing strategy and tactics in projectimplementation. Sloan ManagementReview, 29(1), 33–41.

Smith, M. (2003) Business processdesign: Correlates of success and fail-ure. The Quality Management Journal,10(2), 38–49.

Taylor, S., & Morris, P. (2005).Implementing best practices for the21st century. AACE InternationalTransactions, PM18.1-18.5.

Turner, J. (1999). The handbook of


PHILIPPE BYOSIERE is a professor at Doshisha University, Kyoto, Japan. He received his masters and doctoral degrees

in organizational psychology at the University of Michigan - Ann Arbor. His research interests are in the areas of innovation

management, knowledge creation and competitive advantage, stress in organizations, cross-cultural organizational

research, and organizational behavior.

DENISE J. LUETHGE is a research associate at Doshisha University (ITEC) and an associate professor of corporate strategy

and international business at the University of Michigan – Flint. She holds an MBA from Michigan State University and PhD

in international business from Indiana University-Bloomington. Her research is in the areas of knowledge creation, strategic

orientation in transitional economies, global strategic human resource issues in cross-border acquisitions, cross-cultural

business communication, and pedagogical issues in business simulations and gaming.

project-based management: Improving theprocesses for achieving strategic objectives(2nd ed., vol. 1). London: McGraw-HillPublishing Co.

Urli, B., & Urli, D. (2000).Project management in North

America: Stability of the concepts.Project Management Journal, 31(3),33–43.

Weiser, M., & Morrison, J. (1998).Project memory: Information manage-ment for project teams. Journal of

Management Information Systems, 14(4),149–166.

Wessel, J. (1993). The strategichuman resource management processin practice. Planning Review, 21(5),37–38.

SOFT SKILLS QUANTIFICATION (SSQ) FORPROJECT MANAGER COMPETENCIES

“Soft” skills, also referred to as “micro-

social” skills, are universally recognized

as being critical to successful project

management. Methods of measuring

those skills, however, are to date largely

subjective and nonsystematic. This paper

examines multiple approaches to the

problem, as well as the next evolution in

soft skill assessment: a model that allows

soft skill quantification (SSQ), and its util-

ity in predicting performance. The results

of an SSQ assessment were compared to

formal employee performance results and

individual program manager observa-

tions. Using a small initial sample, the

SSQ model provided statistically verifi-

able alignment with professional project

management employee performance.

Moreover, the assessment identified spe-

cific strengths and weaknesses beyond

what the manager was able to deduce

from observation, and it highlighted the

importance of skill areas that were not

initially thought to be important by the

manager.

Keywords: soft skills; competencies;

quantification; leadership


Vol. 38, No. 2, 30-38, ISSN 8756-9728/03

Introduction and Background

It has long been recognized that screening for and developing nontechnical“soft” skills in project managers and other employees is critical for the con-tinued success of any complex, fast-changing organization. Such skills corre-

spond primarily to A Guide to the Project Management Body of Knowledge (PMBOK®Guide) (PMI, 2004) Knowledge Area entitled Human Resource Management, andsecondarily to the Communications Management Knowledge Area. TheseKnowledge Areas have been demonstrated over the years to be essential to thesuccessful execution of projects across industry and type, thus they are includedin the PMI® certification exam. Recognition of the importance of these factors isnot novel, yet techniques for addressing them are often highly subjective and lacka quantitative foundation.

Recognized Need

The general management literature indicates the importance of leadership andother such “soft” skills in management of organizations and in improving effi-ciency and output, as well as the importance of teamwork and the ability of amanager to create an environment in which the team can flourish (Turner &Muller, 2005; Thamhain, 2004). One technology company anecdotally indicatesthat 90–95% of its documented performance issues annually are concerned withsuch “soft” skills as leadership, teamwork, management, and communication,with only the remaining 5–10% concerned with an employee’s technical knowl-edge or ability (Behavioral Technology, 1999).

This type of conclusion is upheld in the literature. In 1994, in a novel look atthe need for effective interactive skill at the individual level, Wheatley proposed amodel for human organization based on particle physics, in which individualinteractions form a flexible flow toward an end state, and rigid procedures hinderrather than help. The need for a balanced approach between preplanned out-comes and flexible, loose, adaptable tools and management practices clearlyapplies in the project management space as well (Noguiera & Raz, 2006; Pavlack,2004). These types of approaches require that the individuals executing them behighly skilled in interaction, influencing, and stress management.

The need for these skills goes beyond organizational flexibility. Both effectiveteam function and specific manifestations of the project manager management

ABSTRACT

EDWARD MUZIO, Group Harmonics, Inc.DEBORAH J. FISHER, University of New MexicoERV R. THOMAS, Intel CorporationVALERIE PETERS, University of New Mexico

30 JU N E 2007 PR O J E C T MA N A G E M E N T JO U R N A L


• Personal/social, with a number ofsubgroupings (Goleman, Boyatzis,& McKee, 2002)

• Lingusitic/logical-mathematical/spa-tial/bodily-kinesthetic/musical/inter-personal/intrapersonal (Armstrong,1999).

Measurement: The Missing Link

Although the literature is clear on theneed for these micro-social skill areas,it is substantially less clear regardinghow they should be assessed or meas-ured. In a brief review of PMI literatureover the past two years, no workappeared regarding quantitative meas-urement of micro-social or “soft” skillsas they relate to project management.This is perhaps because it is easier tomeasure quantitative technical abilitiesin scheduling, value, contract/procure-ment vocabulary, etc., than to measurethe properties of human social abilities.

The broader literature reviewuncovered a similar trend: a number offrameworks but few common meas-urement metrics. Comparisons madebetween soft skills and performanceshow a linkage, but assessments areoften based upon simple self-reportingprocesses rather than validated analyt-ical measurements (Cote & Miners,2006; Sy, Tram, & O’Hara, 2006). Inany case, none are presented in theproject management space. The impor-tance of these areas, however, is clear.Thus, over the years many ways toaddress them have been devised.

Behavioral Interviewing Method

One popular approach toward deter-mining soft skills is called behavioralinterviewing (BI). This method relieson carefully worded, experience-basedquestions asked by a well-trainedinterviewer. This method is basedupon the axiom that past performanceis a better predictor of future perform-ance than any hypothetical situationalquestion could be. In this method,interviewees cite prior examples ofthemselves, demonstrating key softskill areas, and are at once guided andassessed by the interviewer(Behavioral Technologies, 1999).Although this method exists to pro-vide valuable insight into a candi-

style have been directly linked to proj-ect success (Thamhain, 2004; Turner &Muller, 2005). Efficacy of communica-tion has been found to be critical inmost phases of a project (Hyvari,2006). Transformational leadership, orthe ability of leaders to mobilize andengage their followers at all levels ofmotivation, behavior, aspiration, andneed (Homrig, 2006) while attendingflexibly and proactively to issues asthey arise (Bass, Avolio, Jung, &Berson, 2003), has been widely recog-nized as critically important to projectmanagement (Prabhakar, 2005).

Outside of project management,multiple studies have shown that softskills can compensate for lack of moretraditional cognitive intelligence (Cote& Miners, 2006), and that they oftenare the differentiating factor betweenadequate and stellar performance(Goleman, 2000). Indeed, in the busi-ness bestseller Good to Great (Collins,2001), the author and researchers con-sider the need at the whole-organiza-tional level, taking it a step further.They suggest that the formation of anorganization’s true mission and goalsshould follow the process of hiring thebest performers and getting them intothe proper roles, rather than precedingit. From tactics to organizational strat-egy, the need for so-called “soft skills”has been clearly established.

Descriptive Models

In addition to the need for soft ormicro-social skills, there is a volumi-nous body of written material, both inthe business and psychological disci-plines, about their expression andcharacterization. In 2004, Kendra andTaplin characterized “micro-social”success factors in the literature, whichincluded those areas commonlyreferred to as “soft” skills. Leadership,team building, planning, and manag-ing tasks were among. Although thisexact grouping is unique, like onesnowflake among many, it is far frombeing the only one. Some of the otherproposed groupings in current litera-ture include the following:• Interpersonal/intrapersonal/adaptabil-

ity/stress management/general mood(Butler & Chinowsky, 2006)

date’s likelihood for success, it hasthree critical drawbacks when it isused alone.

The first drawback is that BI reliesupon individual interviews between acandidate and an interviewer. Theaccuracy of the results, therefore, isreduced by human factors. These sub-jective factors are unavoidable on bothsides. In particular, the interviewer’sadeptness at the process is a key factorin the results. Secondly, on the candi-date’s side of the table, adeptness at theBI process comes into play. BI is a well-known process, especially in the worldof university graduates. The candidatewho has rehearsed more and deliversmore smoothly will undoubtedly havean advantage, regardless of their actuallevel of micro-social skill. Finally, BI isa single-point process that happensonly during the screening process. Thefeedback from the interview is rarelyshared with the employee, and thusrarely used to assist with post-hiredevelopment. Moreover, in caseswhere multiple candidates are vyingfor the same position, the candidateultimately selected for the positionusually is one with strong scores acrossthe board, much like the other mem-bers of the team. From a developmentstandpoint, therefore, the areas exam-ined will have been threshold require-ments for the job, but not necessarilyareas in which the employee will beable to make a distinguished contribu-tion to the group.

Self-Assessment Method

A complement to BI that could be con-sidered is self-assessment. In a “role-competency-task” education andtraining model developed for SandiaNational Laboratories (SNL), 35 SNLproject managers were asked to assessthemselves regarding their knowledgeand gap areas (Fisher, Schluter, &Toleti, 2005). The starting point for thedevelopment of the “Sandia ProjectManagement Career DevelopmentAcademy” (SPMCDA) model was thenine PMBOK® Guide KnowledgeAreas. In this study, only five of the35 project managers voluntarilyoffered to take an online exam basedon PMI’s PMP certification test. These


Development of these profession-als is mostly employee-driven.Program management and engineerstogether refer to employee-generateddevelopment plans designed to drivenot only skill development, but alsofuture assignment goals for each engi-neer. Once those goals are established,it is up to the engineer to network withthe appropriate managers and groupmembers to secure the desired assign-ment(s). With frequent job changesand targeted development, both drivenby the employee, the program is amicrocosm of the larger business cli-mate in the United States, in whichthere are about as many job seekers onthe market who are employed as thereare who are unemployed (Goleman,2000). Both accurate candidate screen-ing and effective employee develop-ment are essential in this environment,both for the individual employees andfor the program.

In this pilot study, the qualitativeassessment tool was examined for twopurposes: first, as a possible enhance-ment to the interview and screeningprocess, to give the hiring managersmore information about prospectiveemployees’ abilities before hiringthem; second, as a development plan-ning tool, to give employees and man-agers objective, independent feedbackregarding strengths that should beleveraged and weaknesses that shouldbe addressed.

Initially, 20 engineers, all reportingto the same program manager, wereselected as participants in this pilotstudy. Figure 1 shows the methodologyused in the pilot study. All were giventhe SSQ assessment. Assessments wereprovided directly to the employees,

five project managers were the mostconfident of the 35, and most werealready PMP-certified. Findings regard-ing knowledge gaps from the five proj-ect managers aligned almost identicallywith the self-assessment gaps for all 35project managers (Fisher et al., 2005).However, many studies have shownthat highly successful professionals areby definition those who are extremelyaware of their own weaknesses andactively compensate for them (Nilsen &Campbell, 1993). This alignmentcould decrease substantially if tested ona more representative population thatincluded all levels of professionals.

Success Profile Competency Set

One model that both defines soft skillsand links them clearly to businessobjectives is the “success profile com-petency set,” which defines the follow-ing categories (Fisher et al., 2005):• Core values (company values, profes-

sional, and technical expertise)• Leadership (dealing with change,

taking responsibility, decision-mak-ing, developing self and others)

• Mission success (results, strategicperspective, innovation, businessacumen, customer focus)

•Interpersonal skills (diversity and inclu-sion, building relationships, collabora-tion and teamwork, communication).

Individual developmental plan-ning can be readily linked to these cat-egories as well as to desired businessresults. But again, although this frame-work defines the areas clearly, it doesnot provide a directly quantifiablemeasurement tool to assess them.

SSQ Model

The focus of this research is the nextevolution beyond the aforementioned,more subjective, traditional methods.A desirable future state would be tohave the ability to numerically assessthe micro-social skill areas, therebyproviding soft skills quantification(SSQ). SSQ would provide a new levelof understanding to advance the fieldof leadership assessment. A potentialmodel for SSQ, recently made avail-able, has demonstrated construct andconcurrent validity in initial studies

(Axiometrics International, Inc.,2002). Neither this nor any otherapproach to SSQ has yet been appliedto project management.

In this pilot study, the SSQ modelwas tested on a group of technicalengineering professionals engaged inproject management. Although thegroup for this initial pilot study wassmall (14 individuals), the results werequite compelling. Results of the SSQmodel confirmed job performance rat-ings conducted independently by thesubjects’ management, with high sta-tistical confidence. There was also astrong indication that the SSQ modelmore accurately identified specificmicro-social strengths and weaknessesof the subjects than what their manag-er’s observations predicted.

Methodology

The target population for this pilotstudy was an organization of engineer-ing professionals engaged in projectwork with a standard four- to six-month timeline. This program, whichhires between 40 and 150 recentlygraduated engineers into the parentcompany each year, rotates its employ-ees among two or three different engi-neering organizations in their first yearof work, and then assists them with afinal placement into a more perma-nent position. While on the temporaryassignments, the engineers are givenproject work to be completed withinthe scope of their “rotation.” One indi-vidual, the program manager, overseesthe hiring, development, and place-ment of all of the engineers, and servesas their direct manager during the timethey are engaged in these four- to six-month projects.

Figure 1: Research process flow chart


who took them individually at differ-ent times over approximately a two-week period, all without managerintervention (see Figure 1, step 4). Theexercise involved sorting two represen-tative lists of 18 statements, each inpreferential order. These statementsrepresent various Axiological patternsof thought that have been correlated tospecific soft skill indices (Smith, 2003).To the best knowledge of all involved,neither employees nor the programmanager had any prior experience withthe assessment tool or its underlyingresearch. Neither employee nor manag-er was allowed to see any report untilall had completed the assessment.Basic, standard instructions were com-municated regarding the mechanics ofhow to complete the assessment, butno additional information pertainingto the tool or the research was commu-nicated in advance of all employeesand the manager completing theirassessments.

Prior to the initiation of the assess-ment process just described, theresearchers were provided with “360-degree”-type formal performance feed-back for each employee (see Figure 1,

step 1). The program manager also pro-vided both general information aboutthe program as an organization withperformance requirements, and specificinformation about each of the employ-ees included in the pilot study. The pro-gram manager also assisted in theselection of the six most important softskill areas necessary for employee suc-cess, from the list of those measured bythe instrument (see Table 1 and Figure1, step 2). The areas selected wereresults orientation, interpersonal skills,personal accountability, flexibility,problem solving, and planning andorganization. Those SSQ areas are here-after referred to as the “preselected keycharacteristics.”

Once the preselected key character-istics were determined, the programmanager worked alone to compileinformation regarding the individualsin the pilot study. Detailed informationon each of the 20 employees was sum-marized based upon the program man-ager’s own observations of the staff (seeFigure 1, step 3).

In summary, the information usedfor comparison to the assessmentresults included the following:

• Performance data: Data for eachemployee, ranking from first to last,based upon 360-degree feedback fromcustomers, peers, and management.

• Manager observation: Employeestrengths and weaknesses as observedby the individual manager, includingthe following:

– An observation score by themanager for each employee oneach of the six preselected keycharacteristics, using the follow-ing scale: 1 (not observed); 2(observed to a low level); 3(observed to a moderate level); 4(observed to a high level)

– The one characteristic, from thefull list of 23 (see Table 1),which was expected to be eachemployee’s strongest area

– The one which was expected tobe each employee’s weakest area.

A sample of all available data is sum-marized in Table 2. As indicated inFigure 1, steps 1-3, both formal 360-degree performance data and individ-ual manager observational data werecompiled before the program managerhad received any information about

Table 1: Full list of characteristics measured by the SSQ model (pre-selected key characteristics are shown in bold)


than the BHP population on each ofthe various axes of measurement previ-ously listed. These hypotheses weretested against the null hypothesis thatthe two groups showed no significantdifference. For each measurementitem, a confidence level was calculatedfrom the p-value, representing the con-fidence that the THPs had a statistical-ly higher mean than the BHPs.

The results from the SSQ werecompelling. Four of the six preselectedkey characteristics measured by theSSQ showed a significant differencebetween THP and BHP populationswith greater than 94% confidence. Theother two characteristics had the sameresult with confidence scores above87%. On the other hand, the results ofthe manager’s observational ratings oneach scale were less pronounced. Onlytwo of the six had confidence levelsgreater than 85%; two more had con-fidence levels between 80% and 85%,and the remaining two had confi-

ferences in the SSQ assessment resultsand/or manager observation results ofTHP versus BHP populations.

Top-Half Versus Bottom-Half

Performer Analysis

Table 3 shows a sample of the dataavailable for analysis. This data includ-ed, for each THP and BHP respondent:• A numerical score from the SSQ

model for each of the preselected keycharacteristics (from 1 to 10 to thenearest 0.1)

• A numerical score from the SSQmodel for all other measured charac-teristics

• The manager’s numerical score foreach employee’s skill level for each ofthe preselected key characteristics(from 1 to 4, integer only).

Using this data, multiple one-sided T-tests were used, each testingthe hypothesis that the THP popula-tion had a statistically higher mean

how the assessment measures the char-acteristics, before he had seen theresults of any of his employees’ assess-ments, and before he had taken theassessment himself. As a result, all thisinformation can be safely deemed tobe independent information, distinctfrom the conclusions generated fromthe SSQ assessments themselves.

Data Analysis

From an initial sample of 20 employ-ees, the final sample was reduced to 14usable individuals because some indi-viduals not completing the assessmentsatisfactorily within the deadline.From the formal 360-type perform-ance data on these 14 employees, twosubgroups were defined, the top-halfperformers (THP) and the bottom-halfperformers (BHP). These groupings arebased purely on the company’s per-formance assessment process. Analysisthen focused on determining whetherthere were statistically significant dif-

Table 2: Sample of employee performance data and manager observational data (note: individual identification codes shuffled to preserve confidentiality)


dence levels less than 75%. Figure 2shows these results. It was clear thatthe SSQ model is measuring some-thing ultimately related to how wellthe employee performs. Moreover, themeasurement it is making is moreclosely related to performance thanthe observations made by the employ-ees’ manager.

In addition to the preselected keycharacteristics, other micro-social areasmeasured by the assessment were eval-uated using the same one-sided T-teststo compare the THP and BHP means.Although there were no manager-gen-erated observational pre-ranks forthese areas, the results as measured bythe assessment were again compelling.Four additional skill areas, all of whichseemed on the surface to have rele-vance to the success of a programemployee, were found to demonstratea statistical difference in mean scores(at 94% or greater confidence)between the THP and BHP groupsusing the criteria previously described.These are also shown in Figure 2.

In summary, a total of 10 micro-social areas expected to have relevance

to the performance of a rotation engi-neer ultimately demonstrated a signifi-cant statistical difference in assessmentscores between THP and BHP perform-ers, all with greater than 85% statisticalconfidence. Eight of the 10 were foundto have a confidence level above 93%.By contrast, of the manager’s observa-tions of six of those areas, only twoshowed confidence levels above 93%,while the remaining four showed con-fidence levels below 85%, includingone in the low 70s and one at only50% confidence. Although it is unfor-tunate that manager’s observations didnot include all 10 areas, it is still clearthat on a percentage basis, the assess-ment tool’s evaluation of the skills ofthe employees was more closely relat-ed to their ultimate performance levelthan the manager’s observations.

This pilot study did not assesswhether the specific characteristics aretruly the ones being measured (forexample, whether the assessment’sscore for “flexibility” is actually meas-uring flexibility, or some other charac-teristic); however, prior work on thistopic has already been done (Smith,2003). It is significant that all the

micro-social scales thought in advanceto be important to job performanceultimately showed a relationship toperformance in one or both of the testspreviously mentioned. It is also note-worthy that the subject program in thispilot study intentionally recruits anddevelops employees to produce indi-vidual technical professionals whoperform at a higher level than theirpeers who did not experience this pro-gram. This suggests another question:Would the results be similar or differ-ent in a group with a more diverse per-formance profile? All of thisinformation suggests the need for fur-ther investigation of the SSQ modelwith a larger data sample.

Manager Observations

Before seeing the results of the staffassessments, the program managerreviewed a list all of the behavioralareas that the assessment would meas-ure (see Table 1), and provided foreach employee expectations of theemployee’s strongest area and weakestarea, based upon observation (seeTable 2). These results were also com-pared with the SSQ model results.

Figure 2: Confidence results: SSQ measurements and manager observations vs. actual performance


has little insight into the details of anemployee’s performance so long asobjectives are being met; it is only whendifficulties arise that management isbrought in at a higher level of detail.This is a known flaw of standard man-agement practice (Buckingham &Clifton, 2001).

In the case of this pilot study, themanager’s low success rate in predict-ing the specific strengths of his staff asmeasured by the assessment, coupledwith the close tie between assessmentresults and performance, suggest thatthe SSQ assessment could indeed beused to supplement or even directefforts at employee development. It isalso worth noting that the managerwas not successful in predicting both astrength and a weakness for any singleindividual. If “employee development”truly comprehends both strengths and

Because the assessment measures 23separate areas (see Table 1), the man-ager was considered to have predictedcorrectly if his prediction fell withinthe nearest five areas. In other words, if the manager’s observation of“strongest area” is in the top five meas-ured by the assessment, it is consideredan accurate prediction. Similarly, if themanager’s observation of “weakestarea” is in the bottom five as measuredby the assessment, it is considered tobe accurate. Table 4 summarizes themanager’s success rate.

The conclusion in this section isclearly dependent upon one’s con-clusions from the statistical analysisin the previous section. Given thedata showing that the assessment ismeasuring something that is valu-able and related to employee per-formance, this data suggests someimportant conclusions.

First, the manager had a cumula-tive success rate of only 25% in pre-dicting what the assessment wouldfind (see Table 4). This does not meanthat he was wholly unable to deter-mine his staff’s strengths and weak-nesses, but it does suggest that theassessment may be giving him insightinto areas of skill or shortcoming thathe may not have considered for hisemployees. It is a common problemof management that the manager

does not discover the developmentneeds of the staff until they can beobserved failing.

Even more interestingly, althoughthe manager predicted a weaknessfound by the assessment in 35.7% ofthe cases, he predicted a strengthfound by the assessment less than halfas frequently, at 14.3% (see Table 4).Moreover, his correct assessments ofstrength happened exclusively withinthe THP population, while the majori-ty of his correct assessments of weak-ness (all but one) happened within theBHP population.

This is logical when one considersthat in managing technical, educatedprofessionals, weaknesses are muchmore observable than strengths to amanager, and they are much moreobservable in lower-performing employ-ees. When all is going well, the manager

Table 3: Sample of SSQ data available for analysis

Table 4: Manager’s predictions of strength and weakness vs. SSQ model results


weaknesses, then it is quite possiblethat the assessment provided themanager with new developmentalinformation for every single memberof his organization.

It is worth mentioning here thatthis particular manager recognizesthis challenge and supports the needfor a tool to assist in the area of iden-tifying development areas, bothstrengths and weaknesses, foremployees. That was, in fact, the ini-tial driver in investigating the use ofassessment tools to establish skill lev-els in specific skill areas.

Conclusions

SSQ results and manager observa-tions were analyzed for alignmentwith real-life employee performanceas measured by the company’s formal360-degree employee assessmentprocess. First, the alignment of real-life employee performance with SSQwas statistically stronger than it waswith manager observations. Secondly,the manager was not successful inidentifying specific strong and weakskill areas for individual employees.The manager was particularly blindto the strengths of weaker performersand the weaknesses of stronger per-formers, both of which are essentialfor employee development. Finally,the SSQ instrument highlighted addi-tional skill areas pertinent to per-formance that were not initiallyidentified by the manager. In thisstudy, the SSQ proved to be a valuabletool in characterizing and potentiallyimproving employee performance.

Future Recommendations

The data gathered in this pilot studysupports the conclusion that the SSQmodel measures characteristics relat-ed to performance, and could be usedas a supplemental tool for a managerto identify specific strengths andweaknesses of employees. There is thestrong suggestion that this tool couldprovide valuable, performance-enhancing insight to others engagedin technical work, in this and otherhigh-output corporations. As it turns

out, this SSQ model has been provento be nondiscriminatory across U.S.Equal Opportunity EmploymentCommission areas (race, gender, age,etc.) (Smith, 2003). Future studiesare recommended, and they couldbe longitudinal as well as larger insize, encompassing the use of the model both as a developmenttool, and as a part of a pre-hirescreening strategy.

Acknowledgments

The authors would like to thankIntel Corporation for its coopera-tion in providing information andemployees for studies referencedhere. Also, the authors are gratefulto Innermetrix, Incorporated andTarget Training, International foruse of the SSQ model.

References

Armstrong, Thomas. (1999). 7kinds of smart. New York: PlumePublishing.

Axiometrics International Inc.(2002). Validity studies of theHartman profile model. Copyright2002, Axiometrics International, Inc.Accessed August 2004 at www.axio-metricsinternational.com/AiValidityReport.pdf

Bass, Bernard M., Avolio, BruceJ., Jung, Dong I., & Berson, Yair.(2003). Predicting unit performanceby assessing transformational andtransactional leadership. Journal ofApplied Psychology, 88(2), 207–218.

Behavioral Technology. (1999).Behavioral interviewing participantmanual. Behavioral Technology®, Inc.

Buckingham, Marcus, & Clifton,Donald O. (2001). Now discover yourstrengths. The Free Press.

Butler, Colleen J., & Chinowsky,Paul S. (2006). Emotional intelli-gence and leadership behavior inconstruction executives. Journal ofManagement in Engineering, ASCE,July 2006.

Collins, Jim. (2001). Good togreat. New York: Harper Collins.

Cote, S., & Miners, C. T. H.(2006). Emotional intelligence, cog-

nitive intelligence, and job perform-ance. Administrative Science Quarterly,51(1), 1–28.

Fisher, D. J., Schluter, Lynne, &Toleti, Pavan. (2005, Aug.). Projectmanagement education and trainingprocess for career development.Journal of Construction Engineering andManagement, 131(8), 903–910.

Goleman, Daniel. (2000).Working with emotional intelligence.Bantam Publishing.

Goleman, Daniel, Boyatzis,Richard, & McKee, Annie. (2002).Primal leadership. Boston: HarvardBusiness School.

Homrig, Colonel Mark A. (2006).Transformational leadership. USAirforce Center for Strategic LeadershipStudies. Accessed October 2006 athttp://leadership.au.af.mil/docu-ments/homrig.htm

Hyvari, Irja. (2006). Success ofprojects in different organizationalconditions. Project ManagementJournal, 37(4).

Kendra, Korin, & Taplin, Laura J.(2004). Project success: A culturalframework. Project ManagementJournal, 35(1) 30–45.

Nilsen, Dianne, & Campbell,David P. (1993). Self-observer ratingdiscrepancies: Once an overrater,always an overrater? Human ResourceManager, Summer/Fall.

Nogueira, Juan C., & Raz, Tzvi.(2006). Structure and flexibility ofproject teams under turbulent envi-ronments: An application of agentbases simulation. Project ManagementJournal, 37(2).

Pavlack, Alex. (2004). Projecttroubleshooting: Tiger teams for reac-tive risk management. ProjectManagement Journal, 35(4), 5–14.

Project Management Institute(PMI). (2004). A guide to the projectmanagement body of knowledge(PMBOK® guide). Newtown Square,PA: PMI.

Prabhakar, Guru Prakash.(2005). Switch leadership in proj-ects: An empirical study reflectingthe importance of transformationalleadership on project success across


(2006). Relation of employee andmanager emotional intelligence to jobsatisfaction and performance. Journalof Vocational Behavior, 68(3), 461–473.

Turner, J. Rodney, & Muller, Ralf.(2005). The project manager’s lead-ership style as a success factor onprojects: A literature review. Project

Management Journal, 36(2), 49–61.Thamhain, H. (2004). Team lead-

ership effectiveness in technology-based project environments. ProjectManagement Journal, 35(4) 35–46.

Wheatley, Margaret J. (1994).Leadership and the new science. SanFrancisco: Berrett-Koehler, Publishers.

twenty-eight nations. ProjectManagement Journal, 36(2), 53–60.

Smith, Robert Kinsel. (1998).Hartman validation compendium. ClearDirection, Inc. Accessed August 2004 atwww.cleardirection.com/docs/valida-tion/validationcompendium.pdf

Sy, T., Tram, S., & O’Hara, L.

EDWARD MUZIO received his degree from Cornell University in mechanical engineering, and moved quickly into project

and program management. In the past decade he has led manufacturing infrastructure initiatives and leadership

development programs for companies such as Intel Corporation and the Sematech technology consortium. He has

authored both technical and business-oriented articles, and holds a technical patent. Offered enrollment in the

Massachusetts Institute of Technology Leaders for Manufacturing program, he instead chose to start his own firm; He is

founding CEO of Group Harmonics Inc. and a leader in the application of analytical solutions to the maximization of human

performance.

DEBORAH J. FISHER, PhD, received her ME in civil engineering from Texas A&M University and her PhD in civil engineering

from the University of Texas at Austin. She has more than 30 years of engineering experience, half of which were spent in

the private sector (petrochemical) and. the other half in research and academia in the public sector. Dr. Fisher served as

the director of the Engineering Management Program in the department of industrial engineering at the University of

Houston. She was also the AGC chaired professor at the University of New Mexico, where she currently serves as a visiting

research professor in the department of civil engineering.

ERV R. THOMAS graduated from UC Davis with a degree in mechanical engineering. He worked as a lead design engineer

at Aerospace Corporation before moving to Intel Corporation. Though his Intel career began in the purely technical space,

he has he found his niche recruiting and developing the top engineering students in the U.S., and has spent more than

seven years in that capacity. Thomas also founded and directed a non-profit organization to help teenagers at risk. Overall,

Thomas has invested more than 20,000 hours in finding ways to help professionals, gifted students, and those at risk to

live up to their full potential and deliver outstanding results.

VALERIE PETERS holds an undergraduate degree from Berkeley in industrial engineering, and a master’s degree in

statistics from the University of New Mexico with a focus on business applications. She has developed a variety of technical

and organizational behavior models for companies such as Applied Materials, Intel Corporation, and Sandia National

Laboratories. In addition to being a developer, she has managed worldwide deployment groups to ensure that the tools

she developed functioned correctly. She has served as a developer, coordinator, and trainer for both technical and

nontechnical seminars, and remains active in a variety of organizations including the Society of Women Engineers.


PROJECT SCHEDULING:IMPROVED APPROACH TO INCORPORATEUNCERTAINTY USING BAYESIAN NETWORKS

Project scheduling inevitably involves

uncertainty. The basic inputs (i.e., time,

cost, and resources for each activity) are

not deterministic and are affected by var-

ious sources of uncertainty. Moreover,

there is a causal relationship between

these uncertainty sources and project

parameters; this causality is not modeled

in current state-of-the-art project plan-

ning techniques (such as simulation tech-

niques). This paper introduces an

approach, using Bayesian network mod-

eling, that addresses both uncertainty

and causality in project scheduling.

Bayesian networks have been widely used

in a range of decision-support applica-

tions, but the application to project man-

agement is novel. The model presented

empowers the traditional critical path

method (CPM) to handle uncertainty and

also provides explanatory analysis to elic-

it, represent, and manage different

sources of uncertainty in project planning.

Keywords: project scheduling;

uncertainty; Bayesian networks;

critical path method; CPM


Vol. 38, No. 2, 39-49, ISSN 8756-9728/03

Introduction

Project scheduling is difficult because it inevitably involves uncertainty.Uncertainty in real-world projects arises from the following characteristics:

• Uniqueness (no similar experience)• Variability (trade-off between performance measures like time, cost, and quality) • Ambiguity (lack of clarity, lack of data, lack of structure, and bias in estimates).

Many different techniques and tools have been developed to support betterproject scheduling, and these tools are used seriously by a large majority of proj-ect managers (Fox & Spence, 1998; Pollack-Johnson, 1998). Yet, quantifyinguncertainty is rarely prominent in these approaches.

This paper focuses especially on the problem of handling uncertainty in proj-ect scheduling. The next section elaborates on the nature of uncertainty in projectscheduling and summarizes the current state of the art. The proposed approach isto adapt one of the best-used scheduling techniques, critical path method (CPM)(Kelly, 1961), and incorporate it into an explicit uncertainty model (usingBayesian networks). The paper summarizes the basic CPM methodology and nota-tion, presents a brief introduction to Bayesian networks, and describes how theCPM approach can be incorporated (using a simple illustrative example). Also dis-cussed is a mechanism to implement the model in real-world projects, and sug-gestions on how to move forward and possible future modifications are presented.

The Nature of Uncertainty in Project Scheduling

A Guide to the Project Management Body of Knowledge (PMBOK® Guide)—Third edi-tion (PMI, 2004) identifies risk management as a key area of project management:

“Project risk management includes the processes concerned with conductingrisk management planning, identification, analysis, response, and monitoringand control on a project.”

Central to risk management is the issue of handling uncertainty. Ward andChapman (2003) argued that current project risk management processes induce arestricted focus on managing project uncertainty. They believe it is because theterm “risk” has become associated with “events” rather than more general sourcesof significant uncertainty.

VAHID KHODAKARAMI, Queen Mary University of London, United KingdomNORMAN FENTON, Queen Mary University of London, United KingdomMARTIN NEIL, Queen Mary University of London, United Kingdom

ABSTRACT


pendence of activity duration in aproject network. Moreover, beingevent-oriented (assuming projectrisks as “independent events”),MCS and the tools that implementit do not identify the sources ofuncertainty.

As argued by Ward and Chapman(2003), managing uncertainty in proj-ects is not just about managing per-ceived threats, opportunities, and theirimplication. A proper uncertaintymanagement provides for identifyingvarious sources of uncertainty, under-standing the origins of them, and thenmanaging them to deal with desirableor undesirable implications.

Capturing uncertainty in proj-ects “needs to go beyond variabilityand available data. It needs toaddress ambiguity and incorporatestructure and knowledge” (Chapman& Ward, 2000). In order to measureand analyze uncertainty properly, weneed to model relations betweentrigger (source), and risk and impacts(consequences). Because projects areusually one-off experiences, theiruncertainty is epistemic (i.e., relatedto a lack of complete knowledge)rather than aleatoric (i.e., related torandomness). The duration of a taskis uncertain because there is no sim-ilar experience before, so data isincomplete and suffers from impreci-sion and inaccuracy. The estimationof this sort of uncertainty is mostlysubjective and based on estimatorjudgment. Any estimation is condi-tionally dependent on some assump-tions and conditions—even if they are not mentioned explicitly.These assumptions and conditions are major sources of uncertainty and need to be addressed and han-dled explicitly.

The most well-establishedapproach to handling uncertainty inthese circumstances is the Bayesianapproach (Efron, 2004; Goldstein,2006). Where complex causal rela-tionships are involved, the Bayesianapproach is extended by usingBayesian networks. The challenge isto incorporate the CPM approachinto Bayesian networks.

In different project managementprocesses there are different aspects ofuncertainty. The focus of this paper is onuncertainty in project scheduling. Themost obvious area of uncertainty here isin estimating duration for a particularactivity. Difficulty in this estimation canarise from a lack of knowledge of what isinvolved as well as from the uncertainconsequences of potential threats oropportunities. This uncertainty arisesfrom one or more of the following:• Level of available and required

resources • Trade-off between resources and time• Possible occurrence of uncertain

events (i.e., risks) • Causal factors and interdependencies

including common casual factorsthat affect more than one activity(such as organizational issues)

• Lack of previous experience and use ofsubjective rather than objective data

• Incomplete or imprecise data or lackof data at all

• Uncertainty about the basis of subjec-tive estimation (i.e., bias in estimation).

The best-known technique to sup-port project scheduling is CPM. Thistechnique, which is adapted by themost widely used project managementsoftware tools, is purely deterministic.It makes no attempt to handle or quan-tify uncertainty. However, a number oftechniques, such as program evaluationand review technique (PERT), criticalchain scheduling (CCS) and MonteCarlo simulation (MCS), do try to han-dle uncertainty, as follows:•PERT (Malcom, Roseboom, Clark, &Fazer, 1959; Miller, 1962; Moder,1988) incorporates uncertainty in arestricted sense by using a probabil-ity distribution for each task.Instead of having a single determin-istic value, three different estimates(pessimistic, optimistic, and mostlikely) are approximated. Then the“critical path” and the start and fin-ish date are calculated by the use ofdistributions’ means and applyingprobability rules. Results in PERTare more realistic than CPM, butPERT does not address explicitly anyof the sources of uncertainty previ-ously listed.

• Critical chain (CC) scheduling isbased on Goldratt’s theory of con-straints (Goldratt, 1997). For mini-mizing the impact of Parkinson’sLaw (jobs expand to fill the allocat-ed time), CC uses a 50% confidenceinterval for each task in projectscheduling. The safety time (remain-ing 50%) associated with each taskis shifted to the end of the criticalchain (longest chain) to form theproject buffer. Although it is claimedthat the CC approach is the mostimportant breakthrough in projectmanagement history, its oversim-plicity is a concern for many compa-nies that do not understand both thestrength and weakness of CC andapply it regardless of their particularand unique circumstances (Pinto,1999). The assumption that all taskdurations are overestimated by a cer-tain factor is questionable. The mainissue is: How does the project man-ager determine the safety time? (Raz,Barnes, & Dvir, 2003). CC relies ona fixed, right-skewed probability foractivities, which may be inappropri-ate (Herroelen & Leus, 2001), and asound estimation of project andactivity duration (and consequentlythe buffer size) is still essential(Trietsch, 2005).

• Monte Carlo simulation (MCS) wasfirst proposed for project schedulingin the early 1960s (Van Slyke, 1963)and implemented in the 1980s(Fishman, 1986). In the 1990s,because of improvements in comput-er technology, MCS rapidly becamethe dominant technique for han-dling uncertainty in project schedul-ing (Cook, 2001). A survey by theProject Management Institute (PMI,1999) showed that nearly 20% ofproject management software pack-ages support MCS. For example,PertMaster (PertMaster, 2006)accepts scheduling data from toolslike MS-Project and Primavera andincorporates MCS to provide projectrisk analysis in time and cost.However, the Monte Carlo approachhas attracted some criticism. VanDorp and Duffey (1999) explainedthe weakness of Monte Carlo simula-tion in assuming statistical inde-


CPM Methodology and Notation

CPM (Moder, 1988) is a deterministictechnique that, by use of a network ofdependencies between tasks and givendeterministic values for task durations,calculates the longest path in the net-work called the “critical path.” Thelength of the ”critical path” is the earli-est time for project completion. Thecritical path can be identified by deter-mining the following parameters foreach activity:

D—durationES—earliest start timeEF—earliest finish timeLS—latest start timeLF—latest finish time.

The earliest start and finish timesof each activity are determined byworking forward through the networkand determining the earliest time atwhich an activity can start and finish,considering its predecessor activities.For each activity j:

ESj = Max [ESi + Di ; over predecessor activities i]

EFj = ESj+ Dj

The latest start and finish times arethe latest times that an activity canstart and finish without delaying theproject and are found by workingbackward through the network. Foreach activity i:

LFi = Min [LFj – Dj ; over successor activities j]

LSi = LFi – Di

The activity’s “total float” (TF)(i.e., the amount that the activity’sduration can be increased withoutincreasing the overall project comple-tion time) is the difference in the latestand earliest finish times of each activi-ty. A critical activity is one with no TFand should receive special attention(delay in a critical activity will delaythe entire project). The critical paththen is the path(s) through the net-work whose activities have minimal TF.

The CPM approach is very simpleand provides very useful and funda-mental information about a projectand its activities’ schedule. However,because of its single-point estimateassumption, it is too simplistic to beused in complex projects. The chal-lenge is to incorporate the inevitableuncertainty.

Proposed BN Solution

Bayesian Networks (BNs) are recog-nized as a mature formalism for han-dling causality and uncertainty(Heckerman, Mamdani, & Wellman,1995). This section provides a briefoverview of BNs and describes a newapproach for scheduling project activi-ties in which CPM parameters (i.e., ES,EF, LS, and LF) are determined in a BN.

Bayesian Networks: An OverviewBayesian networks (also known asbelief networks, causal probabilisticnetworks, causal nets, graphical proba-bility networks, probabilistic cause-

effect models, and probabilistic influ-ence diagrams) provide decision sup-port for a wide range of problemsinvolving uncertainty and probabilisticreasoning. Examples of real-worldapplications can be found inHeckerman et al. (1995), Fenton,Krause, and Neil (2002), and Neil,Fenton, Forey, and Harris (2001). A BNis a directed graph, together with anassociated set of probability tables.The graph consists of nodes and arcs.Figure 1 shows a simple BN that mod-els the cause of delay in a particulartask in a project. The nodes representuncertain variables, which may or maynot be observable. Each node has a setof states (e.g. ”on time” and ”late” for”Subcontract” node). The arcs repre-sent causal or influential relationshipsbetween variables. (e.g., “subcontract”and “staff experience” may cause a“delay in task”). There is a probabilitytable for each node, providing theprobabilities of each state of the vari-able. For variables without parents(called “prior” nodes), the table justcontains the marginal probabilities(e.g., for the subcontract” node P(on-time)=0.95 and P(late)=0.05). This isalso called “prior distribution” thatrepresents the prior belief (state ofknowledge) about the variable. Foreach variable with parents, the proba-bility table has conditional probabili-ties for each combination of theparents’ states (see, for example, theprobability table for a “delay in task”

Figure 1: A Bayesian network contains nodes, arcs and probability table


in Figure 1). This is also called the“likelihood function” that representsthe likelihood of a state of a variablegiven a particular state of its parent.

The main use of BNs is in situa-tions that require statistical inference.In addition to statements about theprobabilities of events, users havesome evidence (i.e., some variablestates or events that have actually beenobserved), and can infer the probabili-ties of other variables, which have notas yet been observed. These observedvalues represent a posterior probabili-ty, and by applying Bayesean rules ineach affected node, users can influenceother BN nodes via propagation, mod-ifying the probability distributions. Forexample, the probability that the taskfinishes on time, with no observation,is 0.855 (see Figure 2a). However if weknow that the subcontractor failed todeliver on time, this probabilityupdates to 0.49 (see Figure 2b).

The key benefits of BNs that makethem highly suitable for the projectplanning domain are that they:• Explicitly quantify uncertainty and model

the causal relation between variables• Enable reasoning from effect to cause as

well as from cause to effect (propaga-tion is both “forward” and “backward”)

• Make it possible to overturn previ-ous beliefs in the light of new data

• Make predictions with incomplete data• Combine subjective and objective data• Enable users to arrive at decisions

that are based on visible auditablereasoning.

BNs, as a tool for decision support,have been deployed in domains rang-ing from medicine to politics. BNspotentially address many of the “uncer-tainty” issues previously discussed. Inparticular, incorporating CPM-stylescheduling into a BN framework makesit possible to properly handle uncer-tainty in project scheduling.

There are numerous commercialtools that enable users to build BNmodels and run the propagation calcu-lations. With such tools it is possible toperform fast propagation in large BNs(with hundreds of nodes). In thispaper, AgenaRisk (2006) was used,since it can model continuous vari-ables (as opposed to just discrete).

BN for Activity DurationFigure 3 shows a prototype BN that theauthors have built to model uncertain-ty sources and their affects on durationof a particular activity. The model con-tains variables that capture the uncer-tain nature of activity duration. “Initialduration estimation” is the first esti-mation of the activity’s duration; it isestimated based on historical data,previous experience, or simply expertjudgment. “Resources” incorporate anyaffecting factor that can increase ordecrease the activity duration. It is aranked node, which for simplicity hereis restricted to three levels: low, aver-age, and high. The level of resourcescan be inferred from so-called “indica-tor” nodes. Hence, the causal link isfrom the “resources” directly to observ-

able indicator values like the “cost,”the experience of available “people”and the level of available “technology.”There are many alternative indicators.An important and novel aspect of thisapproach is to allow the model to beadapted to use whichever indicatorsare available.

The power of this model is betterunderstood by showing the results ofrunning it under various scenarios. It ispossible to enter observations any-where in the model to perform not justpredictions but also many types oftrade-off and explanatory analysis. So,for example, observations for the ini-tial duration estimation and resourcescan be entered and the model willshow the distributions for duration.Figure 4 shows how the distribution ofthe activity duration in which the ini-tial estimation is five days changeswhen the level of its availableresources goes from low to high. (Allthe subsequent figures are outputsfrom the AgenaRisk software.)

Another possible analysis in thismodel is the trade-off analysis betweenduration and resources when there is atime constraint for activity durationand it is interesting to know about thelevel of required resource. For example,consider an activity in which the initialduration is estimated as five days butmust be finished in three days. Figure 5shows the probability distribution ofrequired resources to meet this dura-tion constraint. Note how it is skewedtoward high.

Figure 2: New evidence updates the probability


Figure 5: Level of required “Resources” when there is a constraint on “Duration”

Mapping CPM to BNThe main components of CPM net-works are activities. Activities are linkedtogether to represent dependencies. Inorder to map a CPM network to a BN,it is necessary to first map a singleactivity. Each of the activity parametersare represented as a variable (node) inthe BN.

Figure 6 shows a schematic modelof the BN fragment associated with anactivity. It clearly shows the relationbetween the activity parameters andalso the relation with predecessor andsuccessor activities.

The next step is to define the con-necting link between dependent activi-ties. The forward pass in CPM ismapped as a link between the EF ofeach activity to the ES of the successoractivities. The backward-pass in CPM ismapped as a link between the LS ofeach activity to the LF of the predeces-sor activities.

ExampleThe following illustrates this mappingprocess. The example is deliberatelyvery simple to avoid extra complexityin the BN. How the approach can beused in real-size projects is discussedlater in the paper.

Consider a small project with fiveactivities—A, B, C, D, and E. The activ-ity on arc (AOA) network of the projectis shown in Figure 7.

The results of the CPM calculationare summarized in Table 1. ActivitiesA, C, and E with TF=0 are critical andthe overall project takes 20 days (i.e.,earliest finish of activity E).

Figure 8 shows the full BN repre-sentation of the previous example.Each activity has five associated nodes.Forward pass calculation of CPM isdone through the connection betweenthe ES and EF. Activity A, the first activ-ity of the project, has no predecessor,so its ES is set to zero. Activity A ispredecessor for activities B and C sothe EF of activity A is linked to the ESof activities B and C. The EF of activityB is linked to the ES of its successor,activity D. And finally, the EF of activi-ties C and D are connected to the ES ofactivity E. In fact, the ES of activity E isthe maximum of the EF of activities C

Figure 3: Bayesian network for activity duration

Figure 4: Probability distribution for “duration” (days) changes when the level of “resources” changes


one scenario is to see how changingthe resource level affects the projectcompletion time.

Figure 10 compares the distribu-tions for project completion time asthe level of people’s experiencechanges. When people’s experiencechanges from low to high, the meanof finishing time changes from 22.7days to 19.5 days and the 90% confi-dence interval changes from 26.3days to 22.9 days.

Another useful analysis is whenthere is a constraint on the projectcompletion time and we want toknow how many resources are need-ed. Figure 11 illustrates this trade-offbetween project time and requiredresources. If the project needs to becompleted in 18 days (instead of thebaseline 20 days) then the resourcerequired for activity A most likelymust be high; if the project comple-tion is set to 22, the resource level foractivity A moves significantly in thedirection of low.

The next scenario investigates theimpact of risk in activity A on theproject completion time as it isshown in Figure 12. When there is arisk in activity A, the mean of distri-bution for the project completiontime changes from 19.9 days to 22.6days and the 90% confidence intervalchanges from 22.5 days to 25.3 days.

One important advantage ofBNs is their potential for parameterlearning, which is shown in thenext scenario. Imagine activity Aactually finishes in seven days,even though it was originally esti-mated as five days. Because activityA has taken more time than wasexpected, the level of resources hasprobably not been sufficient.

By entering this observation themodel gives the resource probabilityfor activity A as illustrated in Figure13. This can update the analyst’sbelief about the actual level of avail-able resources.

Assuming both activities A and Euse the same resources (e.g., people),the updated knowledge about thelevel of available resources fromactivity A (which is finished) can beentered as evidence in the resources

and D. The EF of activity E is the earli-est time for project completion time.

The same approach is used forbackward CPM calculations connectingthe LF and LS. Activity E is the last activ-ity of the project and has no successor,so its LF is set to EF. Activity E is succes-sor of activities C and D so the LS ofactivity E is linked to the LF of activitiesC and D. The LS of activity D is linkedto the LF of its predecessor activity B.And finally, the LS of activities B and Care linked to the LF of activity A. The LFof activity A is the minimum of the LSof activities B and C.

For simplicity in this example, it isassumed that activities A and E aremore risky and need more detailedanalysis. For all other activities theuncertainty about duration is expressedsimply by a normal distribution.

ResultsThis section explores different scenar-ios of the BN model in Figure 8. Themain objective is to predict the proj-

ect completion time (i.e., the earliestfinish of E) in such a way that it fullycharacterizes uncertainty.

Suppose the initial estimationof activities’ duration is the same asin Table 1. Suppose the resourcelevel for activities A and E is medi-um. If the earliest start of activity Ais set to zero, the distribution forproject completion is shown inFigure 9a. The distribution’s mean is20 days as was expected from theCPM analysis. However, unlikeCPM, the prediction is not a singlepoint and its variance is 4. Figure 9billustrates the cumulative distribu-tion of finishing time, which showsthe probability of completing theproject before a given time. Forexample, with a probability of 90%the project will finish in 22 days.

In addition to this baseline sce-nario, by entering various evidence(observations) to the model, it is pos-sible to analyze the project schedulefrom different aspects. For example,

Figure 6: Schematic of BN for an activity

Figure 7: CPM network


for activity E (which is not startedyet) and consequently updates theproject completion time. Figure 14shows the distributions of comple-tion time when the level of availableresource of activity E is learned fromthe actual duration of activity A.

Another application of parameterlearning in these models is the abilityto incorporate and learn about bias inestimation. So, if there are severalobservations in which actual taskcompletion times are underestimated,the model learns that this may be due

to bias rather than unforeseen risks,and this information will inform sub-sequent predictions. Work on this typeof application (called dynamic learn-ing), is still in progress and can be apossible way of extending the BN ver-sion of CPM.

Figure 8: Overview of BN for example (1)


Object-Oriented Bayesian

Network (OOBN)

It is clear from Figure 8 that even simpleCPM networks lead to fairly large BNs.In real-sized projects with several activi-ties, constructing the network needs ahuge effort, which is not effective espe-

cially for users without much experiencein BNs. However, this complexity can behandled using the so-called object-ori-ented Bayesian network (OOBN)approach (Koller & Pfeffer, 1997). Thisapproach, analogous to the object-ori-ented programming languages, supports

a natural framework for abstraction andrefinement, which allows complexdomains to be described in terms ofinterrelated objects.

The basic element in OOBN is anobject; an entity with an identity, state,and behavior. An object has a set ofattributes each of which is an object.Each object is assigned to a class.Classes provide the ability to describe ageneral, reusable network that can beused in different instances. A class inOOBN is a BN fragment.

The proposed model has a highlyrepetitive structure and fits the object-oriented framework perfectly. Theinternal parts of the activity subnet(see Figure 6) are encapsulated withinthe activity class as shown in Figure 15.

Table 1: Activities’ time (days) and summary of CPM calculations

Figure 9: Distribution of project completion (days) for main scenario in example (1)

Figure 10: Change in project time distribution (days) when level of people's experience changes


Classes can be used as librariesand combined into a model as needed.By connecting interrelated objects,complex networks with several dozennodes can be constructed easily. Figure16 shows the OOBN model for theexample previously presented.

The OOBN approach can also sig-nificantly improve the performance ofinference in the model. Although a fulldiscussion of the OOBN approach tothis particular problem is beyond thescope of this paper, the key point tonote is that there is an existing mecha-nism (and implementation of it) thatenables the proposed solution to begenuinely “scaled-up” to real-worldprojects. Moreover, research is emerg-

ing to develop the new generation ofBNs tools and algorithms that supportOOBN concept both in constructinglarge-scale models and also in propa-gation aspects.

Conclusions and How to Move Forward

Handling risk and uncertainty isincreasingly seen as a crucial compo-nent of project management and plan-ning. One classic problem is how toincorporate uncertainty in projectscheduling. Despite the availability ofdifferent approaches and tools, thedilemma is still challenging. Most cur-rent techniques for handling risk anduncertainty in project scheduling (sim-ulation-based techniques) are often

Figure 11: Probability of required resource changes when the time constraint changes

Figure 13: Learnt probability distribution “resource” when the actual duration is seven days

Figure 12: The impact of occurring risk in activity A on the project completion time


event-oriented and try to model theimpact of possible “threats” on projectperformance. They ignore the sourceof uncertainty and the causal relationsbetween project parameters. Moreadvanced techniques are required tocapture different aspects of uncertaintyin projects.

This paper has proposed a newapproach that makes it possible to

incorporate risk, uncertainty, andcausality in project scheduling.Specifically, the authors have shownhow a Bayesian network model canbe generated from a project’s CPMnetwork. Part of this process is auto-matic and part involves identifyingspecific risks (which may be commonto many activities) and resource indi-cators. The approach brings the full

weight and power of BN analysis tobear on the problem of project sched-uling. This makes it possible to: • Capture different sources of uncer-

tainty and use them to inform proj-ect scheduling

• Express uncertainty about comple-tion time for each activity and thewhole project with full probabilitydistributions

• Model the trade-off between timeand resources in project activities

• Use ”what-if?” analysis • Learn from data so that predictions

become more relevant and accurate.

The application of the approachwas explained by use of a simpleexample. In order to upscale this toreal projects with many activities theapproach must be extended to usethe so-called object-oriented BNs.There is ongoing work to accommo-date such object-oriented modelingso that building a BN version of aCPM is just as simple as building abasic CPM model.

Other extensions to the workdescribed here include: • Incorporating additional uncertainty

sources in the duration network• Handling dynamic parameter learn-

ing as more information becomesavailable when the project progresses

• Handling common causal risks thataffect more than one activity

• Handling management action whenthe project is behind its plan.

Figure 14: completion time (days) based on learned parameters compare with baseline scenario

Figure 15: OO model for the presented example

Figure 15: Activity class encapsulates internal parts of network


References

AgenaRisk. (2006). Available atwww.agenarisk.com

Chapman, C., & Ward, S. (2000).Estimation and evaluation of uncer-tainty: A minimalist first passapproach. International Journal ofProject Management, 18, 369–383.

Cook, M. S. (2001). Real-worldMonte Carlo analysis. Proceeding ofPMI Annual Seminars and Symposium,Nashville, TN.

Efron, B. (2004). Bayesians, fre-quentists, and scientists. Journal of theAmerican Statistical Association,100(469), 1–5.

Fenton, N. E., Krause, P., & Neil,M. (2002). Software measurement:Uncertainty and causal modeling. IEEESoftware 10(4), 116–122.

Fishman, G. S. (1986). A MonteCarlo sampling plan for estimatingnetwork reliability. Operations Research,34(4), 581–594.

Fox, T. L., & Spence, J. W. (1998).Tools of the trade: A survey of projectmanagement tools. Project ManagementJournal, 29, 20–28.

Goldstein, M. (2006). SubjectiveBayesian analysis: Principle and prac-tice2. Bayesian Analysis, 1(3), 403–420.

Goldratt, E. M. (1997). Criticalchain. Great Barrington: The NorthRiver Press Publishing Corporation.

Heckerman, D., Mamdani, A., &Wellman, M. (1995). Real-world appli-

cations of Bayesian networks. CommACM, 38(3), 25–26.

Herroelen, W., & Leus, R. (2001).On the merits and pitfalls of criticalchain scheduling. Journal of OperationsManagement, 19, 559–577.

Kelly, J. E. (1961). Critical pathplanning and scheduling mathematicalbases. Operations Research, 9, 246–320.

Koller, D., & Pfeffer, A. (1997).Object-oriented Bayesian networks. InD. Geiger & P. Shenoy, (Eds.),Proceedings of the Thirteenth annualConference on Uncertainty in ArtificialIntelligence (UAI-97) (pp. 302–313),San Francisco, CA.

Malcolm, D. G., Roseboom, J. H.,Clark, C. E., & Fazer, W. (1959).Application of a technique for researchand development program evaluation.Operations Research, 7(5), 646–669.

Miller R. W. (1962). How to planand control with PERT. HarvardBusiness Review, 93–104.

Moder, J. (1988). Network tech-niques in project management. ProjectManagement Handbook, 2nd edition(pp. 324–373). New York: VanNostrand Reinhold.

Neil, M., Fenton, N., Forey, S. &Harris, R. (2001). Using Bayesian beliefnetworks to predict the reliability ofmilitary vehicles. IEE Computing andControl Engineering Journal 12(1), 11–20.

Pinto, J. K. (1999). Some con-straints on the theory of constraints:

Taking a critical look at the criticalchain. PM Network, 13(8), 49–51.

Pertmaster. (2006). Project risk soft-ware. Available at www.pertmaster.com

PMI. (1999). Project managementsoftware survey, Newtown Square, PA:Project Management Institute.

PMI. (2004). A guide to the projectmanagement body of knowledge (PMBOK®guide – 3rd edition). Newtown Square, PA:Project Management Institute.

Pollack-Johnson, B. (1998).Project management software usagepatterns and suggested research direc-tions for future development. ProjectManagement Journal, 29, 19–29.

Raz, T., Barnes, R., & Dvir, D.(2003). A critical look at critical chainproject management. ProjectManagement Journal, 34, 24–32.

Trietsch, D. (2005). Why critical pathby any other name would smell less sweet?Project Management Journal, 36, 27–36.

Van Dorp, J. R., & Duffey, M. R.(1999). Modelling statistical depend-ence in risk analysis for project net-works. International Journal ofProduction Economics, 58, 17–29.

Van Slyke, R. M. (1963). MonteCarlo methods and the PERT problem.Operations Research, 11(5), 839–860.

Ward, S., & Chapman, C. (2003).Transforming project risk managementinto project uncertainty management.International Journal of ProjectManagement, 21, 97–105.

MARTIN NEIL is a reader in “systems risk” at the Department of Computer

Science, Queen Mary, University of London, where he teaches decision and

risk analysis and software engineering. He is also a joint founder and chief

technology officer of Agena Ltd., which developed and distributes AgenaRisk,

a software product for modeling risk and uncertainty. His interests cover

Bayesian modeling and/or riskquantification in diverse areas: operational risk

in finance, systems and design reliability, project risk, decision support,

simulation, artificial intelligence and personalization, and statistical learning.

He earned a BSc in mathematics, a PhD in statistics and software metrics and

is a chartered engineer.

VAHID KHODAKARAMI is a PhD student at RADAR

group at Queen Mary University of London. He

earned a BSc in industrial engineering from

Tehran Polytechnic and an MSc in industrial

engineering from Sharif University of Technology

in Iran. He has more than 10 years experience in

both academia and industry. He has also

consulted for several companies in project

management and system design. He earned his

second MSc in information technology from

Queen Mary. His research interests include project

management, project risk management, decision-

making and Bayesian networks.

NORMAN FENTON is a professor of computing at

Queen Mary (London University) and is also chief

executive officer of Agena, a company that

specializes in riskmanagement for critical systems.

At Queen Mary he is the computer science

department director of research and he is the head

of the Risk Assessment and Decision Analysis

Research Group (RADAR). His books and

publications on software metrics, formal methods,

and riskanalysis are widely known in the software

engineering community. His recent work has

focused on causal models (Bayesian nets) for risk

assessment in a wide range of application

domains such as vehicle reliability, embedded

software, transport systems, TV personalization

and financial services. He is a chartered engineer

and chartered mathematician and is a fellow of the

British Computer Society. He is a member of the

editorial board of the Software Quality Journal.

Introductory Comments

The Relevance of this Paper

During the past few years, there has been an increasing tendency to drawattention to the particular challenges posed by complex projects (Williams,1999, Richardson, Tait, Roos, & Lissack, 2005) or by complexity in projects

(Baccarini, 1996; Cicmil, 2003a, 2003b; Cicmil, 2005; Cicmil & Marshall 2005;Sommer & Loch, 2004). The discussion, however, has been somewhat hinderedbecause the issue of theoretical foundations in project management research hasbeen a central point of debate among both practitioner and scholarly communi-ties for quite some time.

The discussion about complexity in projects is simply one strand to the grow-ing concern about the dominance of various versions of control theory, opera-tions research, or systems theory in the studies of projects and projectmanagement that are largely normative and prescriptive in character (e.g., Cicmil,2006; Cooke-Davies, 2004a, 2004b, 2004c; Cooke-Davies & Wolstenholme,1998; Hodgson & Cicmil, 2003, 2007; Melgrati & Damiani, 2002; Thomas, 2000;Williams, 2004).

It may be only one strand, but it is an important one. What is at stake is acomprehensive understanding of what it takes to deliver complex projects suc-cessfully in all fields of human endeavor, and not simply in the field comprisingthe traditional arena from which the disciplines of project management emerged.

This paper, then, seeks to make a significant contribution to the discussionabout complexity by reviewing the major ideas that have emerged in the past fewdecades from the school of academic studies that can be characterized as com-plexity science and assessing their possible relevance to the disciplines of projectmanagement. Particular attention is paid to those ideas that are directly relevantto the social complexity created by and among disparate groups of people whotogether make up the team involved in delivering complex projects. After all, oneof the 2006 Nobel Laureates for Physics said, “I’m convinced that over half thecost of a project is socially (contextually) determined.”1

Before embarking on this review, however, it is helpful to undertake a coupleof short excursions to clear the ground of two potential areas that could give riseto serious misunderstandings as the review progresses. The first of these areas isthe realm of vocabulary, and how words are used to shape our understanding ofthe world we live in.

ABSTRACT

WE’RE NOT IN KANSAS ANYMORE, TOTO:MAPPING THE STRANGE LANDSCAPE OFCOMPLEXITY THEORY, AND ITS RELATIONSHIPTO PROJECT MANAGEMENT

TERRY COOKE-DAVIES, Human Systems International, London, U.K.

SVETLANA CICMIL, University of Vaasa, Vaasa, Finland andBristol Business School, University of the West of England , Bristol, U.K.

LYNN CRAWFORD, Bond University, Australia and ESC Lille, France

KURT RICHARDSON, ISCE Group, Mansfield, MA

Both practitioners and researchers in the field

of project management have referred to prob-

lems caused by complexity or problems of

particular significance to complex projects. In

different scientific disciplines investigations

into the behavior of complex dynamical sys-

temsare revealing insightsthat, taken togeth-

er, amount to a challenge to the prevalent

Cartesian/Newtonian/Enlightenment para-

digm from which the practice of project man-

agement has emerged.

Concepts such as nonlinearity, emer-

gence, self-organization, and radical unpre-

dictability have major implications for the

uncodified paradigm that underpins project

management practice and research. Taken

together, they amount to a complementary

way of thinking and talking about projects

and their management that might shed

new light on intractable problems that

appear to plague certain areas of project

management practice.

One strand within complexity studies that

holds particular promise is complex respon-

sive processes of relating, a means of talking

about how human beings interact and learn

and how their interactions evolve over time

and across space. A new program of

research, of which this paper forms part, will

apply this conceptual framework to the lived

experience of project teams, including execu-

tive sponsors, project managers and project

team members.

Keywords: project management theory;

complexity theory; relationships;

complex projects


Vol. 38, No. 2, 50-61, ISSN 8756-9728/03


1 John Mather of NASA, quoted by Charlie Pellerin in private correspondence with Terry Cooke-Davies.

A Word About VocabularyAny discussion about a concept asbroad as complexity or complex proj-ects is bound to encounter risks inher-ent in the use of language. The term“complexity” itself is in widespreadcommon usage, and all readers of thispaper can be expected to have theirown understanding of what the termmeans. Dictionary definitions are notparticularly helpful, generally referringback to the Latin origins of the term(complexus, from complecti) in whichcom is combined with plectari meaningply or braid. The picture that this sug-gests is generally one of the unavoid-able result of a necessary combining,and does not imply a fault or a failure.

If the aim is to say somethingmeaningful about the nature of theworld we live in and, more particular-ly, about the management of projects,then we must be clear that we use twovery different kinds of words in ourconversation. These are objects andideas (Hacking, 2000).

By ideas, Hacking means concep-tions, concepts, beliefs, attitudes to,and theories about. They can be pri-vate or public, clear or vague but in anycase they are discussed, accepted,worked out, shared or contested.Hacking would distinguish between,for example, the idea of complexity inprojects and the objects such as thebehavior of the people engaged in theplanning and controlling activity oncomplex projects and the artifacts thatthey produce through this behavior.

At the same time, we need toaccept that these two types are notindependent from each other. Theobjects we deliberately create arestrongly linked to the ideas we haveabout existing objects, and objectswe’d like to see in the world. Moreover,new ideas are the result of interactionsbetween ideas and objects. We couldeven go so far as to suggest that objectscannot exist (from the viewpoint ofhumans) without ideas about thosesame objects. We do not want to getinto the relationship between ideasand objects too much in this paperother than to say that traditionally theyhave been seen as either totally sepa-

rate (extreme relativism) or more orless one and the same (extreme real-ism). Complexity thinking tends toregard the relationship between ideasand objects as rather more “complex.”

The short of it, however, is thatresearch that takes ideas and treatsthem as if they were objects (in effectreifying them)—what philosopherswould label a naïve realism—is builton shaky foundations.

Unfortunately in literature aboutthe management of projects, there is atendency in practitioner literature toreify processes, and in literaturederived from organizational theory toreify social groupings and organiza-tional units. This results in a blurringbetween objects and ideas, and a lackof methodological integrity to muchquantitative research.

For the purposes of this paper, theissue is further complicated by the factthat the term “complexity theory” isbecoming widely applied to researchin a remarkable diversity of domains,each of which has specialized charac-teristics that distinguish it to somedegree from the others, but fromwhich a number of recurring deepthemes have emerged that link eachdomain to the others (Axelrod &Cohen 2000). This paper proposes thatthese deep themes emerging from theresearch are of particular interest to thestudy of project management.

The Importance of Paradigms The second area in need of clarity is theparadigm that shapes both projectmanagement practice and projectmanagement research.

The term “paradigm” has becomewidespread in discussions of the phi-losophy of science since it was intro-duced by Kuhn in 1962 in the firstedition of The Structure of ScientificRevolutions as a way of describingachievements that arise when a groupof scientists adopt models from whichspring “particular coherent traditionsof scientific research” (Kuhn, 1996, p.10). The term encompasses broadaspects of scientific practice, includinglaw, theory, application, and instru-mentation together.


As Kuhn explains, “the study ofparadigms … is what mainly preparesthe student for membership of the par-ticular scientific community withwhich he (sic) will later practice.Because he there joins men wholearned the bases of their field fromthe same concrete models, his subse-quent practice will seldom evoke overtdisagreement over fundamentals”(Kuhn, 1996, p. 11).

“Paradigms,” asserts Kuhn, “maybe prior to, more binding, and morecomplete than any set of rules … thatcould be unequivocally extracted fromthem” (Kuhn, 1996, p. 46).

It is Kuhn’s assertion that scientif-ic breakthroughs occur when a groupof scientists starts to undertake scien-tific research employing a differentparadigm than that which underpins“normal science.” To paraphrase Kuhn:We do not learn to view the world dif-ferently because research provides uswith new and different answers; ratherwe view the world in new and differentways, and so do different research.

What Kuhn says about the preva-lence of paradigms in determining sci-entific achievements can also beapplied to the prevalence of paradigmsor world views in other fields ofhuman endeavor. Wenger, for example,showed the importance of situatedlearning in shaping the developmentof practice through the absorption of aparadigm when expert practitionersoperate as a community of practice.(Wenger, 1998).

Project management itself embod-ies a paradigm that is more coherent,more binding and more complete thanthe theory on which it is based, or thanthe achievements of the research thatare increasingly being fed back to itsgrowing number of practitioners. Allattempts to describe the paradigm willinevitably rob it of some aspects of itsrichness, but it has been described asrational (Lundin & Soderholm, 1995),normative (Melgrati & Damiani, 2002;Packendorf, 1995), positivist (Smythet al., 2006; Williams, 2004), andreductionist (Koskela & Howell, 2002).That is not to say, of course, that thereare not practitioners who operate


much more challenging non-linear sys-tems such as the weather (which hadstubbornly refused to yield its secretsto the predominant mechanistic think-ing), which are far more common innature than linear ones are. Indeed, itwas while using a computer to simu-late weather systems in 1960 atMassachusetts Institute of Technologythat meteorologist Edward Lorenz(1963) discovered one importantaspect of how non-linearity affects theweather—the principle of “sensitivedependence on initial conditions.”Using a computer model that wentsome way toward simulating thebehavior of a weather system, Lorenzinterrupted one such computer run,and entered the same starting data ashe had done on an earlier run—onlyto find that after a few days the systemstarted to diverge widely from that ofthe earlier run.

Lorenz’s discovery of how minutechanges can have major and unpre-dictable consequences in non-linearsystems became known as the “butter-fly effect” from the title of a paper thathe presented in December 1979 to theAnnual Meeting of the AmericanAssociation for the Advancement ofScience, “Predictability: Does the flapof a butterfly’s wings in Brazil set off atornado in Texas?” Paradoxically, thefact that weather is not predictable inthe long term does not mean that itsbehavior is impossible to understandor explain. Even though weather sys-tems exhibit unstable behaviors, thesystems themselves are not inherentlyunstable. The British weather, forexample, is recognizably the Britishweather, and not that of the Arctic or ofTexas. The reason for that is explainedby the next landmark.

Strange AttractorsIn his further work, Lorenz (1993) alsocontributed to the development of thissecond noteworthy landmark—Strange Attractors. To understandstrange attractors, it is necessary toknow that in any dynamical system(such as a simple pendulum) you canrepresent the state of the system usinga diagram known as phase space. Inthe case of the pendulum, for example,

beyond these ideals, and it is the intentof the research program, of which thispaper forms a part, to provide bothlanguage and conceptual tools to iden-tify and facilitate this.

This paper is not concerned with asociological explanation of the originsof the paradigm that underpins projectmanagement, but the tenor of all theworks cited so far in this connectionmakes it clear that behind the paradigmlies a mechanistic world view derivingfrom Cartesian philosophy, aNewtonian understanding of the natureof reality, and an Enlightenment episte-mology whereby the nature of theworld we live in will be ultimately com-prehensible through empirical research.This is an important assertion, becausethe nature of the deep themes that areemerging from complexity theory canbe said to amount to nothing less thanan expansion and enrichment of theCartesian/Newtonian/Enlightenmentparadigm from which the practice ofproject management has emerged. If itproves to be correct, then the emergingparadigm may well provide projectmanagement with the breakthroughs inpractice that are being called for in theconduct of “complex” projects.

Complexity Theory—

The New Landscape

Now that the ground has been cleared,so to speak, it is possible to embark onthe promised review of the new land-scape, not so as to present a historical-ly rigorous account of thedevelopment of the emerging para-digm, but rather to provide practition-ers and researchers in the field ofproject management a useful overviewof the more significant components ofthe paradigm, including a briefaccount of when and how they arose.

Complexity theory can be definedbroadly as the study of how order,structure, pattern, and novelty arisefrom extremely complicated, apparent-ly chaotic systems and conversely, howcomplex behavior and structureemerges from simple underlying rules.As such, it includes those earlier fieldsof study that are collectively known aschaos theory. The component ele-ments of the emerging paradigm have

arisen from research conducted in aca-demic fields that include life sciences,physical sciences, and mathematics. Torefer to these component elements asfeatures of a landscape is an appropri-ate metaphor, because it ties in directlywith the concept of “fitness land-scapes,” which is one of the ways thatthe emerging paradigm has influencedevolutionary theory in biology.

The broad temporal and discipli-nary relationships between the land-marks that will be discussed in this paperare shown as a diagram in Figure 1. Thenames associated with each landmarkare frequently not the only people toundertake significant research into theparticular field. They are, however, thenames of people who have been close-ly associated with the landmark con-cept, and thus represent a startingpoint for any readers who are interest-ed enough to want to pursue the topicin greater depth. Figure 1 is a map of aterritory that encompasses about a halfcentury of research in many disciplinescarried out in different countries. Assuch, it is inevitably a gross simplifica-tion—and the actual landscape is farmore interconnected that the diagramimplies. What follows now is a briefexplanation of the key landmarks onthe map (Figure 1).

The “Butterfly” EffectDuring the late 1960s and early 1970s,a group of scientists working in a widerange of disciplines became uneasywith the basic assumptions of linearitythat were used as the basis for muchscience, especially in the physical sci-ences. Many researchers in mathemat-ics and physics had concentrated theirattention on discovering underlyingpatterns of cause and effect that couldbe expressed in terms of linear mathe-matics. For some three centuries, aview of the world as a kind of “clock-work masterpiece” (encapsulated inthe mathematics of Newton andLeibnitz), had allowed scientists tounlock many of its mysteries.

The development of computers,however, with their ever-increasingcomputational power, opened thedoor not only to new technologicalachievements, but also to studies of


a simple diagram showing its position(on the X-axis) and its velocity (on theY-axis) translates the behavior of thesystem into a simple spiral, finally set-tling down in a position of rest at theorigin of the diagram. Although such asimple dynamical system as a pendu-lum requires only a simple two-dimen-sional phase space diagram, it ispossible to represent more complexsystems (such as the weather that wasmentioned earlier) by more complexphase space diagrams.

Working in this medium, DavidRuelle, a Belgian-born mathematicalphysicist, working with Floris Takens, aDutch mathematician, began to devel-op the notion of strange attractors todescribe and explain the patterns ofbehavior that they detected whilestudying turbulence in fluids. Theirfindings were first published in 1971(Ruelle and Takens, 1971), and overthe succeeding years their work gainedincreasing recognition. As they did so,these recurring patterns (for that is

what strange attractors, in effect, are)began to provide an explanation forwhy apparently chaotic systems (suchas the weather that has already beendiscussed) display recurring and quasi-predictable features. This insightopened new possibilities for scientistsstudying the behavior of dynamicalsystems in nature, enabling the surpris-ing discovery that complex systems canfollow a number of qualitatively differ-ent attractors, depending upon initialconditions and external perturba-tions—this is very different from sim-ple deterministic chaos.

FractalsThe term fractal, the next feature of thisproposed landscape, was coined by theFrench mathematician BenoitMandelbrot (1982) to describe irregu-lar shapes that repeat themselves innature. It is a form of algebra dealingwith features of self-similarity (theproperty of certain natural objects torepeat themselves on different scales ofsize), and it helps to explain how com-plex patterns from simple guidelines.Although rooted in algebra, it is fractalgeometry that has found the greatestpopularity, with many books, maga-zines, and posters featuring spectacularpictures of such patterns as theMandelbrot set. Fractal geometry isabout the whole system, not its compo-nent parts, and explains mathematicallyhow it is possible to see the same patternrecurring at both a small scale (such asan individual fern leaf) and at a largerscale (such as the plant as a whole). Italso offers insights into the mathemati-cal nature of strange attractors.

Edge of ChaosPerhaps the feature of the new land-scape that has become the most famil-iar to project managers has emergedfrom the life sciences, and is generallyknown as the edge of chaos, a term fea-tured prominently in a best-sellingbook on the whole emerging field ofcomplexity theory (Lewin, 1993).Studies of the evolution and behaviorof living dynamical systems suggestthat these systems manage to demon-strate elements of both chaotic andorderly behavior, and both computer

Figure 1: A diagram of “landmarks” on the landscape of the emerging paradigm


Prigogine for which he was honoredwith the 1978 Nobel prize. In awardingthe prize, the Nobel Committeeobserved that Prigogine’s work had“fundamentally transformed andrevised the science of thermodynamics.”

Prigogine’s primary interest was ina particular kind of complex dynami-cal systems, which he called dissipativestructures because they are continuallyboth receiving and transmitting ener-gy—thus dissipating it. In order toenable experimental science to beapplied to simply one aspect of thesesystems Prigogine chose to studyBénard instability, which is concernedwith the patterns of convection in flu-ids. Through encasing a thin film ofliquid between two sheets of glass,injecting it with a dye, and thenobserving the ensuing patterns of con-vection when the lower part of the filmwas heated, Prigogine was able todemonstrate that even systems com-posed of inanimate fluids, reachedpoints of irreversible change (knownas bifurcations) where the state of thesystem changed in ways that wereimpossible to predict from microscop-ic considerations alone—not becauseof any inadequacies of information,but simply because the outcome wasinherently unpredictable.

This early work led on to moregeneral studies of dissipative structures(or complex dynamical systems as theyare now more generally known) and toa recognition of the potential thatthese systems have for producingunpredictable behavior. Ultimately, itled Prigogine to ask the question, “Isthe future given, or is it under perpetu-al construction?” (Prigogine, 1997).

Self-Organizing SystemsPrigogine’s work in this field made hima Nobel Laureate, but the work that hestarted has been taken forward enthusi-astically in the study of spontaneousself-organization. Examples of complexdynamical systems that seem capable ofself-organization and exercising choicein a manner that makes them inherent-ly unpredictable include hurricanes, liv-ing cells, and human self-organization.What all these systems have in commonis that they exchange matter and energy,

scientists (through cellular automata—more popularly known as artificial life)and evolutionary biologists have car-ried out pioneering work to under-stand why this might be.

Stuart Kauffman, one of the fore-most biologists working in this field(e.g., Kauffman, 1993), likes to use theanalogy of the different states of waterto illustrate the concept. Water can existin a completely orderly state (ice) or achaotic one (steam), but it is in theintermediate form (liquid) that it offersthe best opportunities for the develop-ment of complex activities. The fact thatthese simple molecules can organizeinto three such qualitatively differentstructures is itself significant, but it isfrom Kauffman’s primary area of study(evolutionary biology) that the conceptof “fitness landscapes” has emerged,offering explanations for the success orfailure of different species as their envi-ronment changes.

Scientists working at the Santa FeInstitute, who did much of the earlywork in this field starting in the early1980s, found that both observationsand computer simulations of antcolonies provide some evidence for thisbalance between order and chaos, andfor the “real” existence of a way of exis-tence at “the edge of chaos.” Ants asindividuals exhibit chaotic tendencies,continually switching between franticactivity and static inactivity. The colonyas a whole, however, exhibits a patternof behavior that is both rhythmic andorderly. Experimental studies haverevealed that the pattern is also affectedby population density, with the natureof activity in the nest changing frommore chaotic to more stable, dependingupon how many ants are in the nest atthe particular time.

Kauffman does not claim undis-puted title to have originated the term;co-claimants include Chris Langton andNorman Packard (Lewin, 1993). Hedoes, however, elevate it to primaryplace as one of the four laws he propos-es as candidates for an explanation ofhow evolutionary activity of competi-tion between species coupled with theirevident interdependence can lead to theco-construction of a biosphere, such asthe one we live in (Kauffman, 2000).

Universality—Patterns and Patterningin the WorldOrder and chaos are not confined tothe life sciences. Anyone who has everbeen kept awake at night by a drippingfaucet will recognize that the timebetween drips can vary both in the vol-ume and frequency of the noise. If thefaucet is tightened slightly, the patternslows down, and if it is loosened, thepattern accelerates. The amount bywhich a faucet has to be tightened inorder to double the time between dripsis known as the period doubling factor.Mitch Feigenbaum (1979) is a physi-cist who, in the 1970s, was working tounderstand this “period doubling”effect (which, incidentally, is a com-mon mechanism to move from sim-plicity to chaos), and in 1975discovered that a particular number(approximately 4.669) is associatedwith period doubling in every field—not just dripping faucets.

Ian Stewart (1996), a mathemati-cian, cites this as simply one exampleof the mathematical regularities suchas π or ? (the golden number), whichseem to crop up time and time again asconstants in various laws of geometryor of nature. He also draws attention tothe fact that in nearly all flowers thenumber of petals is one of numbersthat makes up the Fibonacci series ofnumbers 3,5,8,13,21,34,55,89. Lilies,for example, have three petals, butter-cups have five, many delphiniumshave eight, marigolds have 13, astershave 21, and most daisies have 34, 55,or 89.

Why a man-made construct(mathematics) should be so useful inexplaining so many aspects of the nat-ural world remains one of the greatmysteries of life. The observation thatrepetitive patterns occur in the mostdiverse and unlikely fields, however,has become elevated to the principle ofuniversality—one element in theemerging paradigm.

Dissipative StructuresIf the edge of chaos lies close to theheart of evolutionary biology, a similarrevolution has taken place in the realmof physics, perhaps exemplified mostdramatically in the work of Ilya


and remain far from equilibrium. Thefeedback loops that are contained with-in the systems ensure that rich patternsare produced, and that the system itselfbehaves in its own unique way. Thisproduction of complex behavior fromrelatively simple rule-based behaviorand feedback loops allows such systemsto be simulated on modern high-pow-ered computers. It makes possible thelifelike behavior of computer-generatedimages of fish, birds, or bats for the pur-poses of either entertainment (as infilms such as Batman Returns) or for seri-ous study (see Reynolds, 1987).

EmergenceIf universality is one of the observedcharacteristics of complex dynamicalsystems in many fields of study, a sec-ond characteristic that flows from thestudy of these systems is that of emer-gence. As self-organizing systems goabout their daily business, they are con-stantly exchanging matter and energywith their environment, and this allowsthem to remain in a state that is farfrom equilibrium. That allows sponta-neous behavior to give rise to new pat-terns. Such characteristics allow shoalsto respond effectively to predators, forexample, of organisms to adapt to lifein different climatic conditions fromthose within which they evolved. Indoing so, characteristics and patternsemerge that are different in kind as wellas in degree from the characteristics andpatterns of the constituent componentsof the system.

It is these emergent properties ofliving systems that allows novelty andinnovation and provides a credibleaccount of how diversity and varietyarise in order to allow evolution to hap-pen. It also suggests that when dealingwith complex dynamic systems, there isan element of unpredictability aboutthe future that is pregnant with as-yetundreamt of possibilities. As Kauffman(2000, p. 139) so elegantly puts it, “Theuniverse in its persistent becoming isricher than all our dreamings.”

Complex Adaptive SystemsThe last landmark shown in Figure 1arises quite naturally from the preced-ing two—self-organizing systems and

emergence. The difference betweencomplex adaptive systems and self-organizing systems is that the formerhave the capacity to learn from theirexperience, and thus to embody suc-cessful patterns into their repertoire,although there is actually quite a deeprelationship between self-organizingsystems and complex adaptive systems.Adaptive entities can emerge at highlevels of description in simple self-organizing systems, i.e., adaptive sys-tems are not necessarily self-organizingsystems with something extra thrownin. McMillan (2004, pp. 30 & 31) dis-tinguished between a laser beam,which is a self-organizing system, anda human brain, which is a complexadaptive system.

One specific strand of the study ofcomplex adaptive systems—complexresponsive processes of relating, is dealtwith in more detail later in the paper.But before examining how this particu-lar strand of complexity theory relatesto the study of projects and their man-agement, there is one final concept thatoccupies a significant place in theemerging paradigm. It is one that can-not be located in any of the other land-marks, but is itself an emergentproperty of the new fields of study.

IndeterminacyIn fields of study as varied as physics,biology, mathematics, and even philoso-phy, a characteristic is emerging thatchallenges the fundamental tenets of theCartesian/Newtonian/Enlightenmentparadigm. That characteristic is therecognition of the inherent indetermi-nacy of the future of complex dynamicalsystems, and thus of the physical uni-verse itself.

To the annoyance of Einstein whofamously rejected the idea with thestatement, “God does not play dice,”(see Stewart, 1997 page xi)Heisenberg, Schrödinger, and otherpioneers of quantum theory demon-strated that in the subatomic world,even physical matter contains inherentuncertainty. As was previously shown,Prigogine arrived at similar conclu-sions in thermodynamics, andKauffman in biology. In mathematics,Gödel’s theorem demonstrates that it

is impossible to formulate any theo-rem in mathematics that does not con-tain at least one unprovable assertion.And by no means least, Wittgenstein(1953), as he reflected on and rejectedthe conclusions he reached in Tractatus(Wittgenstein, 1921), concluded that itwas impossible to define the condi-tions that are necessary and sufficientin any lower-order characteristic tofully account for the higher-level defi-nition. For example, one cannot pre-cisely predict the next number in eventhe best-defined mathematical seriesof numbers, until the series is com-plete. This will, of course, never hap-pen if the series is infinite. Theimplications of this emerging para-digm for both science, and for thestudy of project management, are fun-damental, and it is appropriate to spellthem out before turning in detail tocomplex adaptive systems.

Implications of the New Landscape for

Science and for Scientific Research

In a plea for an end to what he refers toas a “post-critical philosophy” Polanyi(1958, p. 74) wrote:

“Backed by a science which sternlyprofesses that ultimately all thingsin the world—including all theachievements of man from theHomeric poems to the Critique ofPure Reason—will somehow beexplained in terms of physics andchemistry, these theories assumethat the path to reality lies invari-ably in representing higher things interms of their baser particulars. Thisis indeed almost universally regard-ed today as the supremely criticalmethod, which resists the flatteringillusions cherished by men in theirnobler faculties.”

The developments describedwould seem to answer Polanyi’s reser-vations, in part at least. Summarizingthe foundations to the scientific com-plexity-systems theories, Auyang(1999, p. 341) writes:

“Science reveals complexity unfold-ing in all dimensions and novel fea-tures emerging at all scales andorganizational levels of the universe.The more we know the more we


ogy this is quite a fashionable theory(but not without its opponents).

• Stability is achieved at the level ofpatterned behavior influencing andsimultaneously being influenced bythe patterns at a higher level ofinteraction and governance (theboundary of strange attractor).

A single complex mathematicalformula describes a pattern that keepson repeating itself up and down thescale. One formula describes lots ofthings. It is the kind of thing thatkeeps mathematicians happy. It isused to describe, rather than to pre-dict. It applies to coastlines, weathersystems, clouds, and so on. Fractalpatterns, even though they exist as amathematical ideal, DO approximatecertain features of the natural world.

Prigogine (1980) developed thetheory of dissipative structures, whichwas his early name for self-organizingsystems, by extending the work ofthermodynamics from closed to opensystems. Instead of a world whereentropy rules and systems are subjectto ongoing deterioration, he showedthat open systems are essentially non-linear, dynamic, and able to transformthemselves into new states of being. Ithas been argued that dissipative struc-tures are the basis to all living systems,including human beings.

• Radical unpredictability related to timeflux and outcomes of complex inter-action among agents over time, i.e.,change at the micro-level of the phe-nomenon. “Paradoxically, weathersystems as a whole are not unstable,although they exhibit unstable behav-iors.” Prigogine expressed the funda-mental question “Is the future given,or is it under perpetual construction?”

Complex Responsive Processes

of Relating

Complex responsive processes of relat-ing (CRPR) is a theoretical conceptwithin the conceptual palette of com-plexity thinking in general and com-plex adaptive systems in particular,which has been introduced and arguedfor by Stacey (2001, 2003) and his co-researchers (Stacey, Griffin, & Shaw,2000; Fonseca, 2002; Griffin, 2002;Shaw, 2002; Streatfield, 2001) on the

become aware of how much we donot know. Gone is the image of aclockwork universe. Equally unten-able are the image of a clockwork sci-ence that claims to comprehend allthe diversity by a single method and asingle set of laws and the clockworkscientists who are absorbed in deduc-ing the consequences of the laws byapplying given algorithms. Scientificresearch is a highly creative activity.Scientific creativity, however, is not ananything-goes arbitrariness. There aregeneral guiding principles, which arediscernable across diverse disciplines.”

By no means would all scientistsworking in the field of complexity theo-ry agree with the implication that deter-minism (the clockwork universe) andindeterminism (scientific creativity) aremutually exclusive. The game-of-life isprobably the best example of a clock-work universe that contains (at higherlevels of reality) seemingly creative enti-ties. It is not a question of replacing onesimplistic philosophy with another—rather it is a recognition of paradoxunderpinning the very nature of reality.

For Project Management and ProjectManagement ResearchIf even pure science is finding the needto become more flexible in its researchmethods while not relapsing into “any-thing goes,” is it perhaps too much tohope that research into projects andtheir management will take account ofthese developments and incorporatemethods that investigate both theobjects (human beings) who worktogether in ways that are labeled “proj-ects,” and the ideas that they find use-ful in doing so?

If this provides the basis to aresearch method that is compatiblewith both critical realism (humanbeings working together and learningtogether as complex adaptive systems)and constructivism (human beingsgenerating ideas about their worktogether), then it will represent amajor step forward for project manage-ment research.

Such a method will incorporateseveral of the ideas that have beenbriefly portrayed.

• Nonlinearity—The Butterfly effect,sensitive dependence on initial con-ditions, strange or multiple attractors,self-organization, adaptive systemsand self-transformation. As has beenpreviously seen a complex dynamicalsystem is any system that has withinitself a capacity to respond to its envi-ronment in more than one way. Itintroduces the idea of choice—and itcannot be simply a mechanical sys-tem. You can do the same thing sev-eral times over and get completelydifferent results; small variations canlead to big changes, while big varia-tions can result in minimal change;the paradox of detailed programmingof a system’s path in advance.

• Emergence—Processual rather thanstructural systems paradigm (not onlygoal-oriented but driven by broader,multiple, heterogeneous and fre-quently conflicting agendas, aspira-tions and values). It is at the heart ofthe process of evolving, adapting andtransforming; E. O. Wilson (1971, p.224) at Harvard considers it as a corefeature of life shared by humans andinsects alike; by working collectively,something novel emerges; the role ofcooperation and self-organization isseen as essential for evolution; (seenext the notion of joint action pro-posed by Stacey (2001) as one of thekey premises of human living).

• Explanation of states of stability andinstability; chaos and equilibrium, orderwithin chaos/fractals/complex patterns;self-organization, adaptation and self-transformation. Although the potentialfor chaos resides in every system, chaos,when it emerges, frequently stays with-in the bounds of its attractor(s): Nopoint or pattern of points is ever repeat-ed, but some form of patterningemerges, rather than randomness.

Life scientists in different areashave noticed that life seems able tobalance order and chaos at a place ofbalance known as the edge of chaos.Observations from both nature andartificial life suggest that the edge ofchaos favors evolutionary adaptation.Individual ants behave in a chaoticfashion, but the colony does not(this is the result of the emergence ofrobust high-level structures). In biol-


basis of the problematic capacity ofother theoretical approaches toaddress complexity and paradox incontemporary organizations. It sug-gests a particular way of speakingabout complexity of organizations,organizing, managing, and knowing. Itemphasizes self-referential, reflexivenature of humans, the essentiallyresponsive and participative nature ofhuman processes of relating, and theradical unpredictability of their evolu-tion and outcomes over time. From amethodological point of view, thisconcept puts emphasis on the interactionamong people in organizations and isconcerned with the question of howthe patterned themes of conversations inlocal situations constitute and aresimultaneously constituted by powerrelations in organizations, and howthe potential transformation of theseconversational patterns can inducechange, trigger learning, and createnew knowledge.

With these assumptions andpropositions, CRPR implies an alterna-tive view on management of organiza-tional arrangements, the methodologyof inquiry, the possibility of control,and the role of the individual and thegroup in these processes. The way inwhich organizing, managing, strategicchange, and complexity areapproached from this perspective res-onates with the theoretical traditionsof Mead’s (1934) relational psycholo-gy, Elias’s (1939) concept of the rela-tional character of human nature, andthe experiences with applications ofchaos and complex systems theory tostudying organizations (Stacey, 2003,Stacey et al., 2001), including boundedinstability, self-organizing propertiesof complex systems, and non-linearbehavior over time. It goes a step fur-ther by acknowledging the advantagesof a processual approach over a sys-temic perspective in understandingcomplex and chaotic patterns of relat-ing among individuals and groups,over time, which simultaneously con-stitute and are constituted by a widerorganizational system. This is wherethe main difference between the con-cept of complex adaptive systems (seeearlier reference to McMillan, 2004)

and the one of complex responsiveprocesses of relating becomes visible.

From the perspective of complexresponsive processes of relating, socialstructures, and individual personalitieslargely emerge without overall inten-tion of an agent in the interactionthrough symbols and gestures. Theargument is that the individual and thesocial structures are emerging at thesame time (Stacey, 2003, p. 319).

Under this theory, organizationis an emergent property of manyindividual human beings interactingtogether through their complex respon-sive processes of relating centered onthe use of language simultaneouslyfor conversation and to negotiatesocial status and power relationships.Communication by means of evolvedlanguage is a defining characteristicof human beings, distinguishingthem from all other species of animal(Kauffman, 1993). Central to the the-ory is the recognition that communi-cation is a complex process involvingboth the words that are spoken andthe response that they elicit—indeedthe chain of responses that provide thecontext for an individual conversa-tion or an element of it. The CRPRconcept rejects the distinctionbetween the individual and the groupas unhelpful and argues that it ismore useful to think of individualsrelating to each other through thecomplex processes of vocalized andnon-vocalized communication.

The concept of the complexresponsive process of relating under-stands “human intentions, choices,and actions as essential to, as operat-ing within, the dynamic of daily inter-actions between people.” It argues that“organizing is human experience asthe living present, that is, continualinteractions between humans who areall forming intentions, choosing andacting in relation to each other as theygo about their daily work together”(Stacey et al., 2000, p. 187). The jointaction has therefore a spatial and tem-poral dimension of the process oforganizing—it is located in a specificcontext and is oriented toward anunknown future—the future that is seenfrom this perspective as being under

perpetual construction by the move-ment of human action itself.

Interaction is both communicativeand power-relating, which is respon-sive but the outcomes of which are unpre-dictable at the fine level of detail. Thus,in the process of responding in themedium of symbols, artifacts, feelingsand the unconscious, novelty (e.g.,change of power relations, new knowl-edge, complex collaborative and indi-vidual learning, innovation) can becreated or can emerge. It draws atten-tion to self-organizing capacity of theprocesses of communicative relating inthe medium of (sophisticated) tools,technologies, and artifacts, and theway people make sense of their beingtogether through self-reflection on con-flict, power and the “ordinary” (Stacey,2003). The concept of CRPR refocusesattention on the reflexive monitoring ofinteraction by agents/actors, and theradical unpredictability and uncontrol-lability of the outcomes of action andintersubjective relating. As was demon-strated earlier, learning is what distin-guishes complex adaptive systemsfrom self organizing systems: separa-tion, alignment, cohesion; pro-grammed with only the simplest ofrules. As flocking is an emergent prop-erty of essential bird behavior …. Soorganization and knowledge are emer-gent properties of the essential humanbehavior of communicating—of com-plex responsive processes of relating.

And out of this web of complexresponsive processes arise both theemergent properties of organizationand self-identity. Furthermore, humanaction is seen from this perspective asthe process of perpetual reproductionof identity (individual and collective atthe same time) with the potential fortransformation. It is accomplished withthe use of tools, artifacts, technologies,design systems, and other symbolswhere identities form and are simulta-neously being formed by narrative andpropositional themes. There is thus nodistinction of kind, or of logical level,between the individual and the social.The phenomenon being studied ishuman relating, and the individual isthe singular element of this, while thesocial is the plural. Transformation and


continually iterated, self-organizingprocesses of relating, and if strategicdirection and future goals are contin-ually emerging, reflecting evolvingcollective and individual identities,the question becomes what happensto project control and what the role ofproject manager/leader is. This is alsorelevant to the debate about projectfailure/success outlined earlier. Theimplication of this perspective tounderstanding project arrangementsis that projects are not structures butsocial arrangements that have struc-tural properties that actors draw uponin their social interaction. The con-cept of CRPR refocuses attention onthe reflexive monitoring of interac-tion by agents/actors, and the radicalunpredictability and uncontrollabilityof the outcomes of action and inter-subjective relating (Cicmil, 2006;Cicmil, Williams, Thomas, &Hodgson, 2006).

On Organizing, Power, Identity, andStructure/Agency RelationshipFrom this perspective, projects are seenas processes of continuous reproduc-tion and transformation in the ongoingcommunicative interaction betweenpeople in the present, both their formermembers and people in other organiza-tions. Themes and conversational pat-terns of relating form, and aresimultaneously being formed by, powerrelations in project settings. Power istherefore located in the processes ofconversation that are the processes ofrelating, rather than in one individual(project manager, line manager, seniormanager) or a group of stakeholders(the parent organization, executiveboard, suppliers, the client) who some-how manipulate or dominate others. Asa consequence of differentiatingbetween those who are “in” from thosewho are “out,” certain frameworks oftalking are often legitimized and prevail(Cicmil & Marshall, 2005).

Conceptually, it is possible toexplain how actors create the arrange-ments labeled “project” as both themedium and the outcome of theirsocial practices. The concept of CRPRacknowledges the role of power rela-tions, artifacts, and feelings in the

novelty are possible because of theintrinsic diversity of the people interact-ing, which makes reproduction (mem-ory) of past habitual interactionimperfect and people’s choice ofresponses spontaneous to some extent.This draws attention to the issue of radi-cal unpredictability and the future asbeing perpetually constructed in theprocess of human interaction, as theknown-unknown, and implies whatStacey (2003, p. 390) calls a “fundamen-tally paradoxical theory of causality”.

The proponents of this conceptacknowledge anxiety that peopleexperience in these processes of com-municative interaction by whichexisting power relations and knowl-edge are being sustained or newforms constructed, and are interestedin how that anxiety is lived withwhen the outcomes are not pre-dictable to the finest level of detailand complexity is unavoidable.

The CRPR concept views manage-rial practice, skills and competencies in aparticular way. As opposed to norma-tive/rational perspectives and majorityof processual, strategic choice, learn-ing organizations, and knowledgemanagement theories, which take themethodological position of “theobjective observer where the managerstands outside the organizationunderstood as a system and thinks interms of controlling it” (Stacey, 2003,p. 414), the complex responsiveprocesses perspective takes the follow-ing position: The manager is assumedto him/herself be a participant inthese processes of relating, continu-ously engaged in “emergent enquiryinto what they are doing and whatsteps they should take next” (Stacey),and reflexive in thinking about thenature of their own complex processesof relating in their local situation.What Stacey has proposed within theconcept of complex responsiveprocesses is that the agency lies simul-taneously with individuals and thegroup because they form and they areformed by each other through self-organizing experiences of relating.Even the most powerful individual,Stacey claims, is a participant inhuman interaction.

New Directions for Research

The implications of this different“lens” through which to examine whathappens when people work togetheron a project team, are that new insightscould appear in particular in connec-tion with three aspects of project man-agement, namely human action,radical unpredictability, anxiety, andinseparability of thinking and action;organizing, power, identity, and struc-ture/agency relationship; and manage-rial practice, skills and competencies(Cicmil, Cooke-Davies, Crawford, &Richardson, 2006).

Human Action, Radical Unpredictability,Anxiety, and Inseparability of Thinkingand ActionIf we adopt the perspective of com-plex responsive processes of relating,then project arrangements and set-tings can be seen as a particular kindof pattern of interaction between peo-ple. Two qualities of such interactionare emphasized as follows: (1) inter-action is always communication andcommunication always takes place inthe medium of symbols, where sym-bols, including vocal symbols of lan-guage, are tools created, reproduced,and transformed in that interaction;and (2) patterns of interactionbetween human bodies is alwayspower-relating because, in relating toeach other people, are always simulta-neously constraining and enablingeach other’s actions. Therefore, proj-ect structures and forms of work,including any kind of design toolsand plans for project-based organiz-ing, must be seen as forming andsimultaneously being formed in theprocess of interaction between peo-ple, through a thematic patterning ofconversational themes.

This approach sheds new light onhow we might understand what goeson in the context of projects, whichare conventionally seen in systemicterms, where there is a dual relation-ship between, on the one hand, form-ative unfolding of the envisageddesign toward some pre-given motiva-tion such as a project goal, and on theother hand, rationalist individualchoice of action. If organizations are


process of construction and reproduc-tion of social structures and action.Project working then can be viewed asan evolving pattern of individual andcollective identities emerging in theordinary, everyday local interactionand conversations between people. Incomparison with more mainstreamapproaches to projects and project-based organizing (discussed earlier),attention here is refocused on theimportance of local communicativeinteraction in the present, “particular-ly its thematic patterning, its gesture-response structure and its reflection inideologies and power relations”(Stacey, 2003, p. 404).

On Managerial Practice, Skills, and Competencies Resonating Baets’s (2006, p. xxiv)view, the concept of CRPR refocusesattention from management of projectcomplexity to management in complexi-ty. The authors’ study and argumentfor CRPR as a theoretical lens and interpretative framework in proj-ect management research has animportant mission: to inform andinfluence practice by encouragingpracticing project managers to payattention to quality of the patterns ofconversations and relating and theirown participation in these processes;to learn how to live with anxiety(their own and that of others); tocope reflectively with the paradox ofbeing and not being in control of theproject and still enabling joint actionto move the things on and accom-plish a cooperative sophisticatedactivity labeled “project” (Cicmil,2006; Cicmil et al., 2006).

In contrast to the two pervasiveassumptions within the project man-agement mainstream (plannable con-tent and disciplined control), Stacey(2001, 2003) argued for a view thatorganizations exist in order to enablepeople to accomplish the joint actionrequired for human living.

… and it is in such responsiveprocesses of relating that humanbeings accomplish joint action ofany kind. The key feature of allhuman groups, organisations, insti-tutions and societies is this joint

action. Joint action is only possiblebecause complex responsiveprocesses of relating are patterned incoherent, that is, meaningful, ways(Stacey, 2003, p. 389).

Intentional goal-oriented actionsemerge in the conversations of practi-tioners, managers and non-managersalike, at a local level, and those conversa-tions function as patterning, meaningmaking processes. One of the key man-agement skills can therefore be seen as“courage to carry on creatively despitenot knowing and not being in control,with all the anxiety that this brings”(Stacey, 2003, p. 393).

This concept provides someimportant propositions related to man-agement knowledge and competencies,which refocus attention away frommanagerial intervention “from out-side” promoted by the conventionalbody of project management knowl-edge toward the following (Cooke-Davies & Cicmil, 2005): • The ability of practitioners to engage

in these processes of conversationaland power relating, and reflexivity inthinking about one’s own complexprocesses of relating with others inproject situations

• Sensitivity to qualities of conversation-al life—to themes that form, and aresimultaneously being formed by,power-relating; the ability to enable“free-flowing conversation” (breakingthe stuck patterns of intersubjectiverelating in the local context) in order tocreate a scope for novelty and changeto emerge

• Adequate holding of anxiety in copingwith unpredictability and paradox ofoutcomes of individual and group com-plex conversational relating

• Ethical and moral concerns aboutactions, both intuitive and logical,taken while “thinking on one’s feet”and simultaneously “knowing” and“not knowing,” “being” and “notbeing” in control.

The authors of this paper are cur-rently engaged in just such a programof research, and look forward to beingable to report their findings in thisjournal at a future date.

References

Auyang, S. Y. (1999). Foundationsof Complex-System Theories: InEconomics, Evolutionary Biology andStatistical Physics. Cambridge, UK:Cambridge University Press.

Axelrod, R., & Cohen, M. D. (2000).Harnessing complexity. Organizationalimplications of a scientific frontier. BasicBooks Edition. New York: Simon & Schuster.

Baets, W. (2006). Complexity, learn-ing and organizations: A quantum inter-pretation of business. Routledge: London.

Baccarini, D. (1996). The conceptof project complexity—A review.International Journal of ProjectManagement, 14(4), 201–204.

Cicmil, S. (2003a). Knowledge,interaction and project work: The per-spective of complex responsive processesof relating. Presented at the 19th EGOSColloquium, “Project Organizations,Embeddedness and Repositories ofKnowledge.” Copenhagen, Denmark,3–5 July 2003.

Cicmil, S. (2003b). Some implica-tions of the “complex responsiveprocesses” concept for studying IT-projects. Presented at NFF(Scandinavian Academy of Businessand Management) biannual confer-ence, Track: IT Projects – managingchaos and complexity? Reykjavik,Iceland, August 2003.

Cicmil, S. (2005). Participation,reflection, and learning in project envi-ronments—A multiple perspectiveagenda. In Love, P., Irani, Z., & Fong, S.W. (Eds.), Managing knowledge in projectenvironments (pp. 155–180). MA:Butterworth Heinemann.

Cicmil, S. (2006). Understandingproject management practice throughinterpretative and critical research per-spectives. Project Management Journal,37 (2), 27–37.

Cicmil, S., & Hodgson, D. E.(2004). Knowledge, action, and reflec-tion in management education – Thecase of project management. Presentedat the 20th EGOS Colloquium,Ljubljana, Slovenia, July 2004.

Cicmil, S., & Marshall, D. (2005).Insights into collaboration at projectlevel: complexity, social interactionand procurement mechanisms.


into the temporary organization: Newdirections for project managementresearch. Scandinavian Journal ofManagement, 11(4), 319–333.

Polanyi, M. (1958). Personalknowledge. Towards a post-critical philoso-phy. Chicago: University of ChicagoPress.

Prigogine, I. (1980). From Being toBecoming. New York. Freeman

Prigogine, I. (1997). The end of cer-tainty: Time, chaos and the new laws ofnature. New York: The Free Press.

Project Management Institute.(2000). A guide to the project manage-ment body of knowledge. NewtownSquare, PA: Project ManagementInstitute.

Reynolds, C. W. (1987). Flocks,herds and schools: A distributedbehaviour model. Proceedings of SIG-GRAPH “87”. Computer Graphics.

Richardson, K. A., Tait, A., Roos, J.,& Lissack, M. R. (2005). The coherentmanagement of complex projects andthe potential role of group decisionsupport systems. Managing organiza-tional complexity: Philosophy, theory,and application. In Managing theComplex, pp. 433–458. InformationAge Publishing, Inc.

Ruelle, D., & Takens, F. (1971). Onthe nature of turbulence. Communicationsin Mathematical Physics, 20, 167–192.

Shaw, P. (2002). Changing conver-sations in organizations. A complexityapproach to change. London and NewYork: Routledge.

Smyth, H. J., Morris, P. W. G., &Cooke-Davies, T. (2006) UnderstandingProject Management: philosophicaland methodological issues, Paper pre-sented at Euram 2006, May 17-20, BIManagement School, Oslo.

Sommer, C. S., & Loch, C. H.(2004). Selectionism and learning inprojects with complexity and unfore-seeable uncertainty. ManagementScience, 50, 1334–1348.

Stacey, R. D. (2001). Complexresponsive processes in organizations.Learning and knowledge creation.London and New York: Routledge.

Stacey, R. D. (2003). Complexityand group processes. A radically socialunderstanding of individuals. Hove, UK:Brunner-Routledge.

Building Research and Information,33(6), 523–535.

Cicmil, S., Cooke-Davies, T.,Crawford, L., & Richardson, K. (2006).Impact of complexity theory on projectmanagement: Mapping the field ofcomplexity theory, and using one con-cept of complexity as an interpretiveframework in studying projects andproject management practice. First inter-im report on the progress of the PMI fund-ed project, The research team’s database.

Cicmil, S., Williams, T., Thomas,J., & Hodgson, D. (2006,November).Rethinking project man-agement: Researching the actuality ofprojects. International Journal of ProjectManagement: Special Issue on RethinkingProject Management, 24, 675–686.

Cooke-Davies, T. J. (2002).Learning from experience. ProjectManager Today.

Cooke-Davies, T. J. (2004a).Project management maturity models.The Handbook of Managing Projects.New York: Wiley.

Cooke-Davies, T. J. (2004b).Project success. The Handbook ofManaging Projects. New York: Wiley.

Cooke-Davies, T. J. (2004c). De-engineering project management.Presented at IRNOP VI, Turku, Finland.

Cooke-Davies, T. & Cicmil, S. (2005).Complex Responsive Processes of Relating inOrganisations – An approach to social scienceresearch into the actuality of projects EPSRCnetwork. Rethinking project manage-ment, http://www.umist.ac.uk/depart-ments/civil/research/management/rethinkpm/presentations5.htm

Cooke-Davies, T. J., & Wolstenholme,E. F. (1998). Reshaping project manage-ment education and training. ProjectManager Today, X, 10–12.

Elias, N. ([1939] 2000). TheCivilizing Process, Oxford: Blackwell

Feigenbaum, M. J. (1979). TheUniversal Metric Properties ofNonlinear Transformations. J. Stat.Phys. 21, 669-706.

Fonseca, J. (2002). Complexity andinnovation in organizations. London andNew York: Routledge.

Griffin, D. (2002). The emergenceof leadership. Linking self-organizationand ethics. London and New York:Routledge.

Hacking, I. (2000). The social con-struction of what? Cambridge, MA andLondon, England: Harvard UniversityPress.

Hodgson, D. E., & Cicmil, S.(2003). “Setting the standards”: Theconstruction of “the project” as anorganizational object. Presented to the3rd Critical Management StudiesConference at Lancaster University, July2003.

Hodgson, D., & Cicmil, S. (2007).The politics of standards in modernmanagement: Making “the project’ areality.” Journal of Management Studies,44(3), 431–450.

Kauffman, S. A. (1993). The originsof order. Self-organization and selection inevolution. New York: Oxford Universitypress.

Kauffman, S. A. (2000).Investigations. New York: OxfordUniversity Press.

Koskela, L. & Howell, G. (2002).The underlying theory of project man-agement is obsolete, Proceedings of the2002 PMI Conference, Seattle: PMI.

Kuhn, T. S. (1996). The structure ofscientific revolutions, third ed. Chicago:The University of Chicago Press.

Lewin, R. (1993) Complexity: Lifeat the edge of chaos. Phoenix. London

Lorenz, Edward N. (1963).Deterministic nonperiodic flow. J.Atmospheric Sciences, 20, 130–141.

Lorenz, E. (1993). The essence of chaos.Seattle: University of Washington Press

Lundin, R. A., & Soderholm, A.(1995). A theory of the temporaryorganization. Scandinavian Journal ofManagement, 11(4), 437–455.

McMillan, E. (2004). Complexity,organizations and change. London andNew York: Routledge.

Mandelbrot, B. (1982). The FractalGeometry of Nature. 2nd Edition. SanFrancisco: W. H. Freeman.

Mead, G. H. (1934). Mind, self andsociety. Chicago: Chicago UniversityPress.

Melgrati, A., & Damiani, M.(2002). Rethinking the project man-agement framework: New epistemolo-gy, new insights. Proceedings of PMIResearch Conference. Newtown Square,PA: Project Management Institute.

Packendorff, J. (1995). Inquiring


Stacey, R. D., Griffin, D., & Shaw,P. (2000). Complexity and management.Fad or radical challenge to systems think-ing? London and New York: Routledge.

Stewart, I. (1996). Nature’s num-bers. Discovering order and pattern in theUniverse. Phoenix. London

Stewart, I. (1997). Does God PlayDice? 2nd Edn. London: Penguin

Streatfield, P. J. (2001). The para-dox of control in organizations London,U.K.: Routledge.

Thomas, J. (2000). Making sense ofproject management. Projects As BusinessConstituents and Guiding Motives (collec-tion of best papers from peer reviewed

IRNOP conference), Lundin, R.,Hartman, F., & Navarre, C. (Eds.).

Wenger, E. (1998). Communities ofpractice. Learning, meaning and identity.Cambridge, U.K.: CambridgeUniversity Press.

Williams, T. M. (1999). The needfor new paradigms for complex proj-ects. International Journal of ProjectManagement, 17(5).

Williams, T. M. (2000). Safety reg-ulation changes during projects: Theuse of system dynamics to quantify theeffects of change. International Journalof Project Management, 18(1).

Williams, T. M. (2004).

Assessing and building on the under-lying theory of project managementin the light of badly over-run proj-ects. Project Management InstituteResearch Conference, London.

Wilson, E.O. (1971). The insectsocieties. Cambridge, MA: BelknapPress of Harvard University.

Wittgenstein, L. ([1953] 1973).Philosophical investigations, 3rd ed.Trans G. E.M. Anscome. New York:Prentice Hall.

Wittgenstein, L. ([1921] 1974).Tractatus Logico-Philosophus. 1stpaperback ed. Trans Pears, D. F., &McGuinness, B. F. London and NewYork: Routledge.

LYNN CRAWFORD is the director of Human Systems International Limited, professor of Project Management, ESC Lille,

France, and director of the Project Management Research Group, University of Technology, Sydney. She is involved in

project management practice, education and research. Through Human Systems, she works with leading corporations

that are developing organizational project management competence by sharing and developing knowledge and best

practices as members of a global system of project management knowledge networks. She is currently involved in two

PMI-funded research projects—Exploring the Role of the Executive Sponsor and The Value of Project Management. Results

of a completed study have been published by PMI in a book titled Project Categorization Systems: Aligning Capability withStrategy for Better Results. Lynn has been leading the development of global standards for project management since the

late 1990s.

KURT A. RICHARDSON is the Associate Director for the ISCE Group and is Director of ISCE Publishing, a recently founded

publishing house that specializes in complexity-related publications. He has a BSc(hons) in Physics (1992), MSc in

Astronautics and Space Engineering (1993) and a PhD in Applied Physics (1996). Kurt's current research interests include

the philosophical implications of assuming that everything we observe is the result of underlying complex processes, the

relationship between structure and function, analytical frameworks for intervention design, and robust methods of

reducing complexity, which have resulted in the publication of over 30 journal papers, book chapters, and 7 books.

TERRY COOKE-DAVIES, founder and executive chairman of Human Systems International, has been a practitioner of both

general and project management since the late 1960s and a consultant to blue-chip organizations for more than 20 years.

With a PhD in project management, a bachelor’s degree in theology, and qualifications in electrical engineering,

management accounting, and counseling, he has worked alongside senior leaders and managers in both the public and

the private sectors, to ensure the delivery of business critical change programs and enhance the quality of leadership. He

is co-author with Paul C. Dinsmore of The Right Projects, Done Right and the recipient of the 2006 Sir Monty Finneston

Award for his outstanding contribution to the development of project management as a vehicle for effective change.

SVETLANA CICMIL, BSc (Civ. Eng), MBA, PhD, is currently a guest professor in industrial management at University of

Vaasa, Finland, and is the director of research in operations and information management at Bristol Business School,

University of the West of England, U.K., where she also lectures in project management at the postgraduate level. Her

specific research interests lie in the areas of complexity theory and sociology of project-based work, project management

education, and the development of project management skills. In collaboration with Dr. Damian Hodgson, she published

an influential book, Making Projects Critical (Palgrave), drawing on a series of distinctive workshops with the same name

with contributions from a range of international researchers.


Previous research has helped us under-

stand the role of program managers in

the matrix organization but little has

been done to provide a framework for

understanding the senior management

role. This research offers a step toward

better understanding that role.

Grounded theory research methods were

used to analyze organizational artifacts

and interview transcripts from an aero-

space business division of a major

Fortune 100 corporation. The findings

asserted that five critical behavior con-

structs, Empowerment, Support,Decision-Making, Flexibility, and

Communications, were instrumental in

matrix organizational performance and

were strongly influenced by senior lead-

ership behavior.

Keywords: matrix; matrix management;

multi-project management; organiza-

tional structure


Vol. 38, No. 2, 62-74, ISSN 8756-9728/03

Matrix Versus Traditional Organizational Structures

T he matrix represents a significant shift away from the fundamental tenets oftraditional organizational-structure theory. First, the matrix disregards unityof command. Instead, individuals report to at least two managers: a func-

tional manager who is responsible for their tools, training, and reassignment butdoes not direct their day-to-day work activity; and a program manager who directsthe day-to-day activities of the team members. Second, the matrix structureencourages integration of labor rather than division of labor. Opportunities forsuccessful matrix organizational performance are often found at the intersectionof different tasks and projects, intersections where miscommunication, misinter-pretation, and technical surprises may emerge. Such challenges are handled suc-cessfully when individuals and teams coordinate directly to integrate their laborrather than coordinating through a management structure. Third, traditional hier-archies are challenged by the matrix structure. Technical experts are viewed muchlike the star athlete and management is viewed much like the coach or managerof the team.

Mintzberg (1993) described the matrix as follows: “highly organic struc-ture, with little formalization of behavior; high horizontal job specializationbased on formal training; a tendency to group the specialists in functional unitsfor housekeeping purposes but to deploy them in small, market-based projectteams to do their work; a reliance on the liaison devices to encourage mutualadjustment, the key coordinating mechanism, within and between these teams;and selective decentralization to and within these teams, which are located atvarious places in the organization and involve various mixtures of line man-agers and staff and operating experts.”

Galbraith (1995) described the matrix in terms of the location of generalmanagement decisions. As he sees it, general managers are tasked to balancebroad, multiple, and shifting demands. They must consider cost versus benefit,opportunity versus risk, and tactics versus strategy, to name only a few.According to Galbraith, no executive can cope with the massive informationflow and rapid change present in some complex business environments. Heasserts the matrix structure, sometimes called the mini-general manager struc-ture, delegates many of those general management decisions to the programlevel—to the program managers.

ABSTRACT

LEADERSHIP BEHAVIORS INMATRIX ENVIRONMENTS

JERRY WELLMAN, PMP, Honeywell International


The matrix drives a shift awayfrom the fundamental tenets of tradi-tional organizational structures, thuscreating unique challenges for the sen-ior leadership team, challenges forwhich their training and experiencemay have ill-prepared them.

Research and theory developmenton the matrix began soon after it firstemerged in practice in the late 1950s(Gaddis, 1959; Hall, 1962). A greatdeal of research has been done at theindividual program level (Archibald,1976; Cleland, 1994). That work hasled to a broad array of tools, tech-niques, and training for programteams and program managers operat-ing within a matrix organization. Onthe other end of the spectrum, quite abit has been written about the historyof matrix and its advantages and disad-vantages (Hill, 1979; Knight, 1978).That work has led to some understand-ing about when to implement a matrix(Galbraith, 1973; Mintzberg, 1979).

Although research has given us aframework to understand the role ofthe program manager in the matrixorganization, very little has been doneto provide a framework for under-standing the unique role of seniormanagement, if indeed their role isunique, in a matrix organization. Thisresearch intends to develop a substan-tive theory of the management behav-iors that reinforce the matrixorganization using the grounded theo-ry “Grounded Theory” should be capi-talized because it is a specific researchmethodology. methodology (Glaser &Strauss, 1999). Grounded theory is aninductive methodology concernedwith understanding the perspective ofthe actors involved and relevant prac-tices. The emergent theory offersinsights that may be translated intouseful models for further research.

Researchers who address structureoften think of management as a familyof generic concepts that are littleimpacted by the unique attributes of aparticular structure. Management sys-tems specialists tend to think of onegeneric hierarchical organizationalstructure. As a result, very little isknown about the interdependenciesbetween specific organizational struc-

tures, management systems, and man-agement behaviors. Only recently haveresearchers begun to explore the man-agement problems of leading and inte-grating matrix organizations (Engwall& Kallqvist, 2001a). The aim of thisstudy is to provide an emergent theoryof matrix-organizational management,providing a foundation for furtherresearch that may test the emergenttheory in other matrix organizationsand in nonmatrix organizations. Suchwork may reveal any unique aspects ofthe role of senior management in amatrix organization.

In today’s business environment,talented employees are encouraged torotate often among business units.Managers and executives find them-selves suddenly immersed in organiza-tion models, market forces, andcultures that they may well have neverexperienced. Too often they attempt toapply traditional generic businessschool concepts that are inappropriateor ineffective for the matrix organiza-tion. Indeed the research literatureoffers little that would enable businessschools to ready students for such situ-ations. This research proposes to take astep toward the integration of manage-ment with matrix organizational theo-ry, building toward a foundation forbetter preparing managers and execu-tives for their roles.

Research Context

This research was conducted in anaerospace business unit within aFortune 100 company (see Figure 1).The business, called Aerospace

Business Division (ABD) in this paper,included 2,000 employees located attwo principal sites within the UnitedStates. ABD provided subsystem solu-tions for missiles, launch vehicles,satellites, the Space Shuttle, and theInternational Space Station. Thus, ABDwas a complex organization dealingwith very powerful customers (the U.S. government and aerospace primecontractors). Sales were derived fromnearly 100 different technology pro-grams being conducted in parallel.Those unique yet parallel programsdrove a need for very high informationprocessing capacity and exerteddynamic pressures for shared resourcesas each program evolved.

Over the past 25 years ABD hadvarying degrees of success, sufferingthrough multi-year periods of lacklus-ter performance as well as periods oftruly industry-leading performance.However, ABD generally performedwell compared to peers for the fewyears before and during this research(see Figures 2 and 3). The measuresof that relatively strong performancewere profit and ROI versus industrypeers. Twenty-five year average aero-space industry profits were 4% andROI was 12%, whereas the averageABD profits were 7% and ROI wasgreater than 24% (per the AerospaceIndustries Association, 2001). ABDled competitors in other measures,including third-party customer satis-faction survey results at or above theirbest-in-class competitors; customer-determined award fee scores above95%; new program win rates above

Figure 1: The context of the ABD study


industry averages; and on-time deliv-ery performance at industry bench-mark levels.

The business unit employed amatrix organizational structure sinceits beginning in the 1950’s. Theresearch findings indicate that at timesthe structure was aggressively support-ed and reinforced by the culture andleadership while at other times thosesame forces thwarted it. However, theform of that structure was unusuallyconsistent. Throughout the period thestructure included functional groups(engineering, production, procure-ment, quality assurance, contracts,etc.) and an independent programmanagement group. Individuals fromthe functional groups were “matrixed”

into a specific program or programs.However, the functional bond was rel-atively stronger or weaker at differentperiods of time and within differentbusiness segments. In spite of this fluc-tuation, the general consensus amongthe leadership team was that the busi-ness unit had operated as a programmanagement matrix throughout itsorganizational life.

The relative stability of the work-force, less than 5% annual employeeturnover, and the projects durations,some as long as 30 years, resulted ina management team that had experi-enced both the ups and downs ofbusiness performance. As a result, ithad developed deep personal beliefsabout the senior leadership behav-

iors and organizational cultures thateither enabled or retarded organiza-tional performance.

Research Methodology

Grounded theory research methodswere adopted for this research for tworeasons. First, the research was aimedat the extension of existing theory.Grounded theory is generally deemedappropriate for such efforts as it allowstheory to emerge from the interviewtranscripts and organization artifacts.Second, the methodology is reputed tohelp separate researcher biases frominterpretation of the data.

A total of 47 program managersand executives were interviewed overa three-month period. The interviewdata were supplemented with anarray of document artifacts fromfinancial data to employee surveyfeedback to customer survey feedbackto periodic reports. The intervieweesranged in age from 38 to 67. Sevenwere female. Two of the executiveswere retirees, having left the firmwithin the preceding five years. Themean industry experience was morethan 21 years and the mean ABDexperience was nearly 20 years. Theprogram managers were generallyvery experienced in that position, themean tenure being about 10 years.Table 1 profiles the interviewees.

The interviews were facilitated viaopen-ended questions intended toencourage individuals to describethose senior leadership and organiza-tional culture attributes that hadeither facilitated business success dur-ing good times or inhibited successduring bad times. The resulting 1,500pages of interview transcripts wereanalyzed to identify recurrent themesand concepts associated with the goodand bad times in the business. Thosethemes and concepts were then testedagainst archive data includingemployee survey results, periodicreports, etc., with an eye to find evi-dence reinforcing or challenging thethemes and concepts. Preliminary con-clusions were then tested via follow-upinterviews with a subset of the original47 individuals. This iterative process(interviews, concepts identification,

Figure 2: ABD had strong customer appreciation

Figure 3: ABD had strong financial performance


Table 1: Interviewee profiles


concepts aggregation, theme analysis,testing against organizational artifacts,review with interviewees, adaptation,repeat) led to the identification of 15 emergent concepts as describedfurther next. Figure 4 shows theresearch methodology.

Emergent Concepts

The initial analysis resulted in the 15concepts listed in Table 2, anddescribed in the following text. Theconcept descriptions are supplementedwith sample verbatim interview state-ments that attempt to enhance andcolor the concept descriptions.

Access captures the notion of amanager’s ability to get time with exec-utive management as desired. Several ofthe program managers made the pointthat having ready access to get advice,share new information, or get a deci-sion was an important organizationalbehavior that made it easier to managetheir programs. Representative tran-script examples include the following:

“I would hate to lose access to on-siteupper management. I would hate tofeel I can’t just pick up the phone andcall them if I need them or walk overto their office. I don’t want to losethat. I also don’t want to lose theautonomy of making the decisions,being able to manage my programs.” “I am thinking of times when I havehad to say to upper management Ireally need your time and I need youto work your schedules to make timefor me. They have always adjustedtheir schedules as necessary.”

Accountability captures the notionof being responsible and required toaccount for one’s actions. The partici-pants addressed accountability from atleast four perspectives. Some talkedabout the expectation that the func-tional groups within the organizationneed to be accountable to the programneeds and expectations. Othersdescribed the concept of distributedaccountability, meaning that different

individuals and groups within theorganization clearly understand andfeel accountable for their particularroles in supporting organization goals.Others described the role that manage-ment review processes play in reinforc-ing accountability. Another perspectivewas the need for accountability butalso flexibility in the manner in whichindividuals or groups meet theirresponsibilities. One transcript exam-ple is offered as follows:

“I think the biggest difference betweenthe two (referring to the “successful” ver-sus ‘unsuccessful’ periods in the organiza-tion’s history) was a willingness toentrust to program management thelevel of responsibility and sense ofownership … the willingness to trustprogram management with theresponsibility to make things happen,and to be able to work effectively withhis (or her) team towards what needsto be done from a program require-ments standpoint. I think the clarityof who is responsible and the claritythat the focal point for the programsresides with the program managerand he knows that he is responsiblefor that. And that is being clearlyunderstood by the organization andsupported by the organizational struc-ture. I think it’s been a very healthyway to operate.”

Active Listening captures the notionthat everyone must be attentive, ask forclarification, and persistently restateunderstandings to assure mutual inter-pretation. The participants seemedfocused on the notion that free and opendialogue intended for mutual under-standing to reach a decision or agree ona plan of action—as opposed to a contestof wills, debating skills, and influencebrokering—was the desired culture. Thisconcept ties closely to the concept ofCommunication and the concept of OpenRelationships, both described later.

“There was enough active listening,and I say enough, it wasn’t just activelistening; there was enough activelistening to understand what wasgoing on and hear from enough peo-ple that we would all hear the samething and all acknowledge what thesituation was and what the pros and

Figure 4: Research methodology

Table 2: Emergent concepts


cons were, and following that therewas a decision. There wasn’t a needfor more data, go back and find outmore, think we need to go talk tosomebody else. So I like the idea thatit wasn’t autocratic kind of decision-making, there was enough active lis-tening that you feel like even if therewas a decision that maybe you didn’tagree with you felt like you had a fairswing at the ball and that was themanagement behavior that I thinkworked real well.”

Allow Mistakes captures the notionthat empowered people will take risksand thus have failures or make mistakes.The participants often expressed anappreciation for the environment inwhich one could take a prudent riskwithout fear of reprisal and recalledtimes when that was not always so.They talked about being mentored sothat they could learn and grow frommistakes. They asserted that removing,or perhaps minimizing, the fear of pun-ishment for mistakes freed them to bemore creative and entrepreneurial. Theactive monitoring and mentoring gavethem a sense of operating within allow-able boundaries without stifling them.

“I have never been ever been pun-ished for (1) breaking a whole bunchof rules, or (2) I guess having some-thing not work out assuming we werereaching for the right objectives.”

Balance refers to continuallyadjusting the organization priorities,resources, and energies to maintain abalance among competing, and oftenconflicting, objectives. The participantscommented on the role of the programmanagement organization, and of theindividual program managers to bal-ance competing interests between oneprogram and another, between pro-grams and functions, between cus-tomers and the organization, betweenprocess discipline and tailoring tomeet unique needs, etc.

“I think management has a big rolein deciding where that balance isgoing to be. You have reviews of allthe program plans, you have reviewsof how we’re going to run the pro-grams, what the tailoring of the dif-

ferent organizations are going to be,you know …. You have responsibil-ity for the overall direction and Ithink that’s where it comes in play.Certainly we have all the forms andthe structures that are set up on thevarious systems that we have that wefollow as templates. So it’s there andavailable for us to use as templates.And then we also can say okay this isnot necessary for this particular pro-gram, this particular tool, this par-ticular application.”

Communication is commonlydefined to be the act of impartinginformation but a second definition isa means of connecting different places,such as a door, passage, or road.Participants certainly described boththese aspects of communication. Asmentioned earlier, the concept of com-munication was raised very often (over80 instances) and by over 80% of theparticipants. A few transcript examplesto give the reader a full appreciationfor the range and depth of this conceptare as follows:

“I am talking about the open con-versation, the ability to feel that as asubordinate you can push back thedirections that don’t make sense,don’t appear right, without any fearof retribution.” “I think that theneed for communication is greateras the complexity goes up. As thecomplexity rises, everything is inter-connected and everybody needs toknow what the others are doing.”

“Okay, well, when I look back to myexperience in (aerospace business otherthan ABD) there just really wasn’t acommunication system there, formalor informal. If I was working on aspecific program, the informationwas closed, and as a contract person Imight not know there was somethinggoing on in the program. That couldcreate an issue on the contract itself,because it had to do with peoplebeing afraid to let the informationout to let other people know therewas something going on. But, it defi-nitely did hinder things because youreally couldn’t work with peoplebecause you didn’t know how to.”

Customer Relationships includesthose relationships between individu-als and groups within the organizationand with individuals and groups in thecustomer community. One attribute ofcomplex organizations is customerintimacy and the participant com-ments that follow demonstrate thatcustomer intimacy was a part of theirwork environment. Their commentssuggest an appreciation for the need,value, and richness of those relation-ships that are woven into the fabric oftheir everyday work. Indeed, every par-ticipant (perhaps excluding the humanresources and information systemsexecutives) in this research had dailycontact with customers, as many asone-third the employee populationhad direct contact with customers sev-eral times a month, and over 80% ofthe population had direct customercontact several times a year.

“Our customers have a lot of inputon how we do our business. I believethey tend to pay us for that. But theydo want to know what is going onand they don’t hesitate to intervenein our processes in various ways.”

“I think that for us the thing thatsets our organization apart … is thatcustomer interface is so important. Ithink that it’s there because the cus-tomers will come … anybody (atany level in the organization). If youdon’t have pretty good flow of infor-mation going back and forth, it canget nasty.”

Decision-Making was identifiedmore than 50 times and occurred inabout one-half of the transcripts.Decision-making was described fromthree perspectives. First, there was adimension of participation. Participantsexpressed a belief that everyone’s viewwould be welcomed, that it was okay todisagree with executives, and that hid-den agendas were frowned upon.Second, there was a dimension of deci-siveness. Program team members, pro-gram managers, and executives wereexpected to make decisions crisply. Eventhough decisiveness seems to conflictwith participation, there were no expres-sions of systemic conflict between the


two. Instead, there seemed to be a per-ception that both timeliness and broadinvolvement were somehow accommo-dated. Third, there was a dimension ofvaluing clear thinking and rational deci-sion-making more than the results ofthe decisions. The emphasis was onmaking sound decisions more than ongetting good outcomes.

“A second level of how decisionsare made and kind of anotherunique feature here, one that Ithink has a lot of value, is with thelocal leadership team. The ability todeal with a problem that potential-ly can affect other functions, forexample, but having the peopleright there, call them together, workthrough the problems and issues,and then coming back with a deci-sion in a fairly quick order I think isanother quick healthy way of howin this organization how decisionscome about and everybody isinformed about it, knows about,there are not surprises, and the deci-sion can be made and then com-municated back to the appropriateparty or parties.”

“Even if the decision wasn’t totalagreement with my way of thinking,I had good rationale and I was ableto say, ‘You know, that’s a goodpoint, I hadn’t really thought aboutit that way.’ And so it wasn’t a mat-ter about saving face and egos andso forth, it was about let’s talkabout it in the open, make a deci-sion and make some consensushere if at all possible.”

Decisiveness denotes recognizingwhen making a decision is moreimportant than gathering more infor-mation and then making that decisionquickly. The continual balancing andrebalancing of resources to respond tonew information is an inherent part oflife in the organization. The intervie-wees, program managers and execu-tives alike, commented often on thebenefits of being decisive.

“A second level of how decisions aremade and kind of another uniquefeature here, one that I think has alot of value, is with the local leader-ship team. They are co-located and

some decisions come in that havespillover effect on engineering, oroperations, or business develop-ment. And with us being togetherit’s quick for us to come togetherand say we have an issue, let’s have aquick meeting and let’s talk aboutthis because you might be impactedand I need to know if this is going tohave any kind of an obstacle to you,we can get together and go workthrough it fairly quickly and comeup with a fairly quick decision,obviously within the latitudes of thedecisions we can make. The abilityto deal with a problem that poten-tially can affect other functions, forexample, but having the peopleright there, call them together, workthrough the problems and issues,and then coming back with a deci-sion in a fairly quick order I think isanother healthy way this organiza-tion makes decisions come aboutand everybody is informed about it,knows about, there are not surpris-es, and the decision can be madeand then communicated back to theappropriate party or parties.”

Empowerment occurred in 87instances as an important attribute atABD. Interviewees clearly perceived theempowering environment to be an ele-ment in ABD’s success.

“The culture has changed a lot in10–15 years …. I think it’s gottenmore, definitely more empower-ment to lower levels, more per-sonable interaction, approachableexecutives, willing to be able to sitdown and have a conversation onan equal level versus—you know,‘Oh gosh, I’m talking to an execu-tive, I’ve got to watch what I say ortemper it.’”

Flexibility was seen as the attributeof bending without breaking, of beingadaptable. Several executives and pro-gram managers alike appreciated thisconcept. Their dialogue addressed theflexibility necessary to meet the uniqueneeds of each program or situation.They talked about the necessity and thevalue of flexibility in decision-making,methodology, and approach. They

generally did not describe the countertension that arises in such a flexibleenvironment, but later discussionwith executives brought out the pointthat internal processes become lessefficient and more complex as theygrow more flexible. They cited the ten-sion between efficient processes andnecessary flexibility as a major leader-ship challenge.

“I’m very much a believer that theformal structure needs to be a guideas opposed to the master. And that Ihave come to appreciate what I per-sonally like to operate in a non-rigidfashion. The structure is importantto provide some direction. And forme personally I like having a struc-ture in place so that you’ve got acommon language but that you’renot a slave to that.”

Open Relationships captures thenotion that participants described asan environment in which opinionsare sought, individuals feel free todisagree with anyone including exec-utives, and arguments are healthy.One person described ABD as havingan environment where “the truthcan’t hide.”

“And contrasting this behavior withwhere I’ve been it encourages infor-mation flow up, unfiltered. I sayunfiltered, people come to meet-ings they want to tell us goodthings. They won’t hide big prob-lems. They’re not going to throwstupid little problems at us they’reworking. They don’t come therelying. They don’t try to hide infor-mation. I think it’s because of ourcollective behavior on how we han-dle bad news.”

“… you don’t fear informationbeing conveyed to the surprise of thefunctional organization of the pro-gram management organization, it iswell understood that before themeeting ever occurs that there is anissue and the issue is going to betalked about and nobody I thinkfeels they have to go protect them-selves in those meetings for a couplereasons. There isn’t any managementin those meetings that you have tolook to be protected from neither.”


Support was often mentioned as animportant expectation by individualsof their upper management. They feltthey both needed and received support,enabling them to accomplish theirgoals in a very dynamic environment.

“I think a lot of the expectations areplaced on ourselves, that we feel likeit’s our role to facilitate, it’s our role tohelp, it’s our role to support, it’s ourrole to take the obstacles away.” (com-ment from an executive staff member)

“I had full management support togo and work that program and anyproblems that came about in thenormal course of doing business.The support was there, and therewas the ability to go for input andconsultation but there wasn’t anymicromanaging applied to it.”

Tools/Processes—Tools includes refer-ences to Integrated Cost/ScheduleControl Systems (IC/SCS), ManufacturingResource Planning (MRP), MicrosoftProject, etc. Processes includes the activitiesinvolved in scheduling, budgeting, track-ing performance, monitoring suppliers,and a myriad of other processes essentialto the successful execution of a complexprogram. Several of the participants didnot distinguish between tools andprocesses or procedures, suggesting theyunderstood the intent of those processesand viewed them as tools rather thanobstacles. Tools were usually described asassets to be employed in whatever uniqueadaptation made sense for the situation,rather than as devices to be used in onlyone way. Thus the concept of flexibilityseems to be intimately connected with theconcept of tools.

“Well, to me the tools are there toallow you to be able to monitor andto run the program and those toolsare there also for management to seethe bigger picture. … It also allowsme the freedom to pick which toolsthat are most effective on the pro-gram. Let me tailor those tools tosome extent needed for a particularprogram. We do everything frommajor production programs here tosmall engineering studies. The sametools can be tailored to be used forthat wide range.”

Trust was described as trust in oneanother, trust in the tools and systems,and trust in the executive staff.

“Let me make sure I’m a little cleareron that ‘and left me alone,’ it wasn’tlike they turned away and didn’t payattention, they paid great attentionand they understood at any point intime through what I was able to con-vey to them, the relative health ofthe program and how we were per-forming and where we thought wewould end up. So by ‘left alone,’ theytrusted that they had prepared meright and that they could allow meto go and demonstrate that I wasgoing to have the skill and the desireto keep doing that.”

These 15 concepts formed a foun-dation for further analyses asdescribed next.

Concept Relationships

The 15 initial emergent relationshipswere further tested and modified orenriched through iterative reviews withseveral of the participants. The result-ing concept groupings and their rela-tionships are depicted in Figure 5.

The approach to identifying therelationships among the conceptsbegan with a search for overlapping orconcurrent concept foundations in thetranscripts. Several transcript segments

were coded to multiple concepts. Forexample, the following paragraph wascoded under four different concepts:Decision-Making, Communication, OpenRelationships, and Active Listening.

“Even if the decision wasn’t in totalagreement with my way of thinking, Ihad good rationale and I was able tosay, ‘You know, that’s a good point, Ihadn’t really thought about it that way.’And so it wasn’t a matter about savingface and egos and so forth, it wasabout let’s talk about it in the open,make a decision and make some con-sensus here if at all possible.”

Thus, some concepts tended to befounded on shared transcript passagesand some concepts tended to be groupedwithin a discussion theme. Shared tran-script text formed the initial basis for ini-tially discovering interrelationships.

These emergent groupings were thentested by a review of organizational arti-facts. For example, a review of statusreports and memos suggested that theorganization practiced empowermentand keeping decisions at the mutuallyagreed levels. Phrases such as “the pro-gram team decided” .and “the two pro-grams resolved the resource conflict by”… were common in the reports andmemos. As another example, the employeesurvey results reinforced the concept thatempowerment was a widely recognizedand appreciated dimension of the culture.

Figure 5: Emergent concepts and relationships


As the concept groupings and rela-tionships began to take shape, theywere tested against common-sense rela-tionships. For example the concepts ofAccountability, Trust, and AllowingMistakes seemed to often arise in asso-ciation with the concept ofEmpowerment. These logical groupingsand relationships were also validatedvia follow-up conversations with select-ed interviewees. For example, during adiscussion with three executives aboutFigure 5, one of them noted thatEmpowerment and Support were culturalphenomena whereas Decision-Makingand Flexibility/Balance might be consid-ered organizational competencies. Theexecutives then collectively decided thedashed line depicting the two dimen-sions should be added to the model.The relationship model was progres-sively tested and enriched through suchiterative feedback with the participants.

The initial review of grouped con-cepts quickly uncovered five majorgroupings and one relatively stand-alone concept. Three concepts,Empowerment, Accountability, andTrust, tended to be addressed in thesame or adjacent transcript excerpts. Afourth concept, Allow Mistakes, didnot occur as often in the transcriptsbut when it did occur it was usually inthe context of either Empowerment orDecision-Making.

Accountability, Trust, and AllowingMistakes were found to be supportivebehaviors or attitudes that enabledmanagers to feel empowered. Thus, theiterative analysis and testing resultedin a decision to group these four con-cepts together under the overarchingconcept of Empowerment.

Three other concepts, Support,Open Relationships, and Access, tendedto be grouped together in the tran-script excerpts. One other concept,Active Listening, had been assigned toseveral pieces of transcript text but thespecific phrase “active listening” wasfound in only three transcript excerpts.It was related to Decision-Making in twoinstances, to Support in one instance,and to both Decision-Making andSupport in another instance. Differentphrasings of the same concept werefound in relationship to Decision-

Making more often than they werefound in relationship to Support. ThusActive-Listening seemed to be tied toboth Decision-Making and to Support,but the frequency of occurrence of rela-tionship and feedback from some ofthe interviewees led the author to affil-iate Active Listening with Decision-Making. Thus, Open Relationships andAccess were found to be behaviors thatenabled program managers and execu-tives to perceive the presence of a sup-portive environment. Indeed,discussions with two of the intervie-wees reinforced the affiliation. Thethree concepts, Support, OpenRelationships, and Access were groupedunder the overarching concept Support.

The three concepts of Decision-Making, Active Listening, andDecisiveness often occurred in commonor adjacent text. Active Listening cap-tures the notion that people seek tounderstand one another’s perspectives.The concept captures a sense of mutu-al respect and a belief that people areworking toward shared objectives.Decision-Making captures a notion oftimeliness in the continual makingand adjusting of decisions that happenso frequently across the matrix organi-zation. Thus, Decision-Making was cho-sen to capture the three concepts underone grouping.

Three concepts, Flexibility,Tools/Processes, and Balance, appeared atfirst to be unrelated concepts but itera-tive transcript analysis and discussionswith some of the executive staff led theauthor to conclude that they wereindeed interrelated in the text and inthe interviewee’s minds. Flexibility wasmost often discussed in regard to thefreedom to adapt tools and processesto meet the needs of a particular pro-gram situation or challenge.

Interviewees described how theywere required to stay within thebounds of guidelines, policies and pro-cedures. They also acknowledged thepresence of tools and processes to helpthem accomplish their work. But, theywere quick to say they felt free to adaptthose tools and processes.

Tools/Processes was originally iden-tified as two separate concepts.However, as the iterative analysis con-

tinued it became clear that that thetools and the processes themselveswere not particularly unique to theinterviewees. They were very muchaware of the flexibility the organiza-tion permitted them in applying thetools and using the processes. Thetools and processes were perceived tohave value because they could be tai-lored. Thus Tools/Processes andFlexibility became paired concepts.

Balance was described with respectto several competing forces, customerneeds versus business unit goals, near-term versus long-term objectives, costversus technical performance, etc.Thus, Balance captures the inherentconflict within any organization, andparticularly within a matrix organiza-tion where competing objectives arerampant. One of the frequentlydescribed conflicts was the conflictbetween the need for discipline andrigor in the use of tools and processesin order to operate the business effi-ciently and cost effectively versus theneed for flexibility to accommodateunique program needs. The executivestaff in particular referred to the needto balance flexibility with efficiencyand control.

Thus, Balance, Tools/Processes, andFlexibility were included under the over-arching concept of Flexibility/Balance.

Until this point in the research,Customer Relations was addressed fre-quently enough to retain it as an inde-pendent concept. As the iterativeanalysis and testing of the conceptrelationships evolved, it became clearthat Customer Relations represented anorganizational sensitivity. That sensi-tivity was a manifestation of the pres-ence of very powerful customers, oneof the basic assumptions for a complexorganization (Galbraith, 1973; Knight,1976; Mintzberg, 1979). Clearly, manyof the interviewees recognized thepresence and influence of those power-ful customers. But, the concept ofCustomer Relations seemed to be dis-tinct from the other concepts and rela-tions previously described. Thus, apreliminary model including the otherconcepts but not Customer Relationswas built then tested through discus-sions with a few of the interviewees.


There appeared to be no reinforcement forincluding Customer Relations. Thus, the con-cept was dropped from further analysis.

Communications was a recurringtheme, mentioned twice as often asany other concept. However,Communications seemed to be affiliatedwith nearly all the other concepts. Noscheme logically allowed the author toseparate this concept fromEmpowerment, Support, Decision-Making, or Flexibility/Balance. In theend it seemed that Communications wasan integrating concept that bridgedmany of the other concepts.Communications fostered the iterativeestablishment of objectives/priorities,the development of plans and alloca-tion of resources, the recognition ofissues. When this relationship wasdescribed to the executives, they saw itas a useful representation of their ownperceptions. Thus, Communicationsappeared to be the “glue” concept thatheld the other concepts together.

The research suggests that fourmetaconcepts, Empowerment, Support,Decision-Making, and Flexibility/Balancewere all important to the culture atABD. The participants seemed to viewthem as prerequisite to enabling suc-cessful execution of the many parallelprograms that make up the business.An effective communication networkenables the organization to continual-ly determine its objectives, establishpriorities among those objectives,develop and execute plans, recognizeobstacles, and adapt its objectivesaccordingly. When Communications iseffective, Empowerment, Support,Flexibility, and Decision-Making are fos-tered. When Communications is ineffec-tive, Empowerment, Support, Flexibility,and Decision-Making are impeded, andthus the organization is impeded.

Research Findings

The findings from this research sup-port many of the assertions and find-ings from previous theorydevelopment in the field of matrixmanagement. Examples include lead-ing a multiple command structure,empowerment of individuals andteams to innovate solutions, seniormanagement coordination rather than

direction, a democratic culture, etc.The findings also build on previoustheory development. For example, thefindings offer a perspective on the rela-tionship between dimensions of cul-ture, structure, and competence.

The notion of an empowering cul-ture is strongly reinforced by thisresearch. Time and again, the interviewtranscripts and organizational artifactsreinforced the notion that ABD is avery empowering organization. In onesurvey of program managers, everyABD respondent indicated they feltempowered to accomplish the pro-gram objectives. The participantsexpressed time and again the impor-tance of empowerment and the extentof empowerment at ABD.

The literature states that a matriximposes significant disadvantages.One disadvantage is that the individu-als working in the organization may beanxious about their job security. Therationale has been that they are fre-quently moved from project to project,at times working for long hours toresolve a problem and at other timestrying to fill the day with useful work.This research did not validate suchassertions. In fact, ABD employee sur-vey results and anecdotal commentssuggest strongly that workforce moraleand enthusiasm was consistently high,even during a three-year period ofbusiness decline. The interviews anddiscussions with the human resourcesstaff attributed the absence of signifi-cant anxiety to a broad engagement bythe workforce in strategic activity, to acommitment by the ABD leadership towork staff reductions through attritionrather than layoffs, and to a continualcommunication campaign. It is recog-nized that a matrix may foster suchanxiety, but ABD appears to have man-aged that risk effectively. Perhaps suchanxiety is not necessarily an unavoid-able attribute of the matrix.

According to Mintzberg (1979),another disadvantage of a matrix isthat all the coordination across somany interfaces imposes a heavy costin the form of meetings and commu-nication. It has been asserted thatmany hours each day are spent defin-ing issues, gathering information, dis-

cussing it, interpreting it, agreeing onits meaning, developing actions, andagreeing on them. This high communi-cation cost means the organization isnot very efficient at doing ordinarythings. It is geared to do the uniquerather than the routine. This researchvalidates such an assertion. Indeed,Communication was by far the mostoften mentioned of any concepts. Theresearch also affirms the assertion thata high cost is paid in the form of manymeetings and discussion activities.

Mintzberg (1979) also argued thatthe culture surrounding a successfulmatrix is open, empowering, and dem-ocratic. He contended that cooperationand teamwork are highly valued. Yet,individual creativity and decisivenessare treasured. Personal satisfactioncomes from creatively accomplishingchallenging projects and overcomingsurprises. Again, this research validatesMintzberg’s position. The interviewtranscripts and the previously conduct-ed employee survey data describe ahighly democratic environment. Infact, the specific words “open” and“empowerment” emerged from theresearch as fundamental concepts.

Mintzberg (1979) also contendedthat a large percentage of the workforceis directly involved in strategy formula-tion or wants to understand the strate-gies. Decisions are usually madethrough drawn-out consensus build-ing. Yet, when the all-too-frequent sur-prise occurs, the appropriate leaders areexpected to be decisive. Because teamaccomplishments are valued over indi-vidual accomplishments, individualsfreely assist one another. Again thisresearch reinforces the points. ABDemployed focused strategy teams as themeans by which a significant percent-age of the workforce, perhaps as manyas 10%, participated in strategy devel-opment and execution. Mintzberg’spoint about decisiveness is reinforcedby this research.

Indeed the ABD organizationexhibits many of the attributesMintzberg describes. The structure androles fit his model quite well. At ABDthere appeared to be a widely heldbelief that rules, policies, and proce-dures are guidelines to be adapted as


necessary to achieve unique programgoals. It seems that ABD had devel-oped a culture that recognizes theinnate conflict between control andinnovation. The culture and organiza-tional practices have become adept atcontinually balancing those opposingforces. The focus on empowermentand support of teams and individualexperts enabled rapid and effectivedecision-making at the most appropri-ate level in the organization.

Hopwood (1978) provided a use-ful model for understanding the com-munications challenge for a matrixorganization. Most of his tenets arereinforced by this research. For exam-ple, he asserts that “The matrix organ-ization is complex and multi-faceted.The effectiveness of the whole organi-zation depends on achieving a contin-uous balance between vertical andlateral processes. … Rather than look-ing for a single integrated and tidyinformation system, it needs to berecognized that to survive and flour-ish the matrix requires a diversity ofinformation systems.” Indeed, ABDused a wide array of such systems toshare information.

But Hopwood (1978) does notaddress an additional communicationchallenge identified in this research. AtABD, there is a strong emphasis thatsuch systems must be adaptable quick-ly to the unique needs of a particularsituation or challenge. The systemsneed to be flexible/adaptable to theunique needs of individual projects.Such flexibility/adaptability is essen-tial to foster the primary reason for thematrix structure—innovation—inno-vation that is only possible if there isample flexibility in all systems. If aproject team is faced with a uniquechallenge or has identified a uniqueapproach, the team may need anequally unique information system todevelop and manage that challenge orapproach. The information systemsmust be able to accommodate, evenencourage, such innovation. Such flex-ibility and adaptability was held atABD to be as valuable as the systemsand tools themselves.

Further, at ABD everyone wassearching for more, better, and fresher

information. Program teams were con-tinually faced with new and emergingchallenges for which they needed tofind solutions. They continually adjust-ed their plans in response to changingneeds and options. Specialists were con-tinually scouring technology sources insearch of solutions to technical chal-lenges. Functional leaders were continu-ally shuffling people and physical assetsbetween projects as the needs changed.Process owners were continually adapt-ing their processes to accommodate theunique and changing project needswhile still keeping the processes undercontrol. Executive management wascontinually monitoring project activityin order to make appropriate prioritydecisions between projects. Thisresearch suggests that matrix communi-cation systems must be not only robustbut also agile to support effective matrixorganizational performance.

Davis and Lawrence (1977) con-tend that the senior leadership role ina matrix is unique. First, the executivemust head up a dual command struc-ture, the functional departments andthe programs. They argued this respon-sibility involves “power balancing,managing the decision context, andstandard setting.” They pointed outthat the executive must maintain anappropriate balance of power betweenthe two organizational chains, con-tending that somewhat equal dualpressures are the crucible for the effec-tive decision-making the matrix isintended to provide. As noted earlier,this research suggests the executivechallenge requires balancing at leasttwo organizational chains and perhapsmore. The ABD leadership wasattempting to balance no fewer thantwo such chains, and may fairly be saidto have been attempting to balance asmany as four or five parallel organiza-tional chains.

Davis and Lawrence (1977) alsoargued that because the matrix mustdeal with a huge flow of informationand many decisions. The executivemust make some of those decisionsbut must also empower others to makeappropriate decisions and must set thecontext and culture for effective deci-sion-making. Indeed, this research

supports such an assertion. Effectivedecision-making appears to be criticalin at least two dimensions. First, thisresearch suggests that empoweringothers and keeping decisions at theappropriate levels in the organizationis a valued behavior. Risks must beembraced and mistakes must be forgiv-en, but people must be capable.Second, the research suggests that it isimportant to appreciate when to con-tinue gathering information and per-spectives versus when to make adecision. This research suggests thatthe leadership team must be decisivein setting priorities and allocating crit-ical resources when such decisionsreach their level. Yet they must foster aculture in which peers at lower levelsin the organization can usually makesuch decisions.

The literature describes culture asan integral part of the matrix (Graham,1997; Johns, 1997). It is asserted thatthose environments best suited for amatrix have several common attrib-utes. Well-educated professionals whoseek challenge and empowermentpopulate these complex high-technol-ogy environments. The long-term inti-mate relationships with powerfulcustomers foster a strong customerfocus and a broader view of the overallmarketplace. But these attributes areunavoidable cultural influences thatmust be acknowledged and dealt with.

The matrix itself exerts equallypowerful influences on the culture.First, ambiguity is a fact of life.Individuals may be assigned to morethan one program, may be movedfrom program to program as required,and may find themselves near the endof a program assignment with no obvi-ous next assignment in sight. There isconfusion about whom their boss is,the program manager or the function-al manager, and whom to impress inorder to get a raise. The continual con-flict over resources and the continualchanging of priorities adds to theambiguity. The leadership mustacknowledge and continually work tocontain such anxiety. Second, theorganization is rife with discussionrather than decisions. Program teamsmeet. Process teams meet. Functional


departments meet. Program membersmeet with customers and with suppli-ers. There are conflict resolution meet-ings, status meetings, problemresolution meetings, etc. It seems thatnothing can be decided without aseries of meetings. This research sup-ports such findings. Indeed, the ABDmatrix displayed all the behaviorscited. The continual meetings anddebates fostered a consensus buildingculture rather than a hero culture. Theorganization worked hard to keepmeetings productive and focused with-out inhibiting creativity.

Implications for Matrix

Organization Leaders

The findings from this research suggestthe following guidance for senior lead-ers in a matrix environment. Some ofthose suggestions are addressed in theprevious section, comparing theresearch finding to the matrix literature.A few others are summarized next.

Leaders must empower the projectmanagers and the functional managersto deal with the continual onslaught ofchange, allowing them the freedom toadapt the resources, policies, andprocesses to meet the needs of themoment. Program teams and teamleaders at ABD were told that theircommitments to customers would behonored by the organization. If thosecommitments were too often inappro-priate, the individuals were counseledor replaced—but the commitmentswere honored nonetheless. But, asMike Kami, and organizational changeconsultant says, “don’t empower dum-mies.” Leaders must put a great deal ofenergy into developing capable andexperienced managers who have thejudgment and maturity to make suchdecisions. Leaders must also foster acontinual sharing of organizationalstrategies and priorities so that thosemanagement decisions can be madewithin an appropriate context.

Leaders must provide support forthe empowered managers. Supporttakes several forms. First, the man-agers’ decisions themselves must besupported. Second-guessing willquickly douse any sense of empower-ment. Decisions that turn out to be

mistakes must be treated as opportuni-ties to learn and improve rather thancritiques of the individual’s judgment.Second, the infrastructure must be flex-ible enough to deal with the uniqueproblems and solutions while alsoremaining in control. Senior leader-ship must instill this flexible/controldynamic in the social interactionbetween program and functionalteams. Third, senior leadership mustmake available the resources needed tosupport the management decisions. Itdoes no good to empower managers todecide to add staff to deal with a crisis,yet deny them the ability to actuallyadd that staff.

The marketplace and businessenvironment in which the matrixorganization is used requires a greatercompetency at decision-making thando other businesses and organizationstructures. The values include team-work, customer sensitivity, diversity,innovation, etc. and are common to allappraisals. The non-value-based por-tion of the appraisal addresses specificgoals, of which about 70% were team-based or shared. The leadership and theculture must value savvy thoughtprocesses and decision-making morethan desired outcomes. We have allseen instances of success in spite of ter-rible decisions and actions, and wehave all seen failure even thought everypossible action was taken to try toachieve success. Managers must believethey are being asked to understand andtake appropriate risks, moving boldlyto influence the outcomes. But theymust also believe that it is not the out-comes that will determine their careers.

Open, dynamic, multi-format,and egalitarian communications is anessential cultural attribute and anessential competency. Individuals atABD were encouraged to share andsolicit information from those whohad it or needed it without regard tothe organizational hierarchy. Teamsthat rapidly collected, spread, digested,and acted on information were praisedlavishly. Senior leaders must foster aperpetual flow of information withinand across teams, within and acrossfunctional groups, between andaround management layers, etc. The

organization must value the effectivedecision-making that emerges fromshared information rather than thepossession of the information itself. Asone interviewee said, “ABD is a placewhere the truth can’t hide.”

Implications for Further Research

This research is a step toward furtherunderstanding the matrix organizationstructure and the management andleadership role in that structure.Similar research in other aerospacebusiness units using a matrix structuremight validate, modify, or challengethe concepts and relationships foundin this research. Similar research inother industries, such as constructionor medicine, might bring a differentperspective to the concepts and rela-tionships. Research in less-successfulbusiness units might validate or chal-lenge the conclusions reached withthis study.

The concepts themselves may befertile ground for future research. Onemight investigate the relative impor-tance of the concepts in influencingmatrix organizational success. Onemight test the relationships betweenconcepts in several organizationalstructures to better understand the rel-ative importance of those relationshipsacross different structures.

References

Archibald, R. D. (1976). Managinghigh-technology programs and projects.New York: John Wiley & Sons.

Engwall, M., & Kallqvest, A. S.(2001a). Dynamics of a multi-projectmatrix: Conflicts and coordination (p.14): Paper published through a web-site at the Fenix ResearchProgram/IMIT.

Engwall, M., & Kallqvist, A. S.(2001b). Exploring the multi-projectmatrix: Process dynamics of a projecti-fied organization. Paper presented atThe Academy of Management—HowGovernment Matters, Washington, D.C.

Gaddis, P. O. (1959). The projectmanager. Harvard Business Review,May-June.

Galbraith, J. (1973). Designingcomplex organizations. Reading, MA:Addison Wesley.


JERRY WELLEMAN, PMP, is vice president of the Military Segment of Honeywell International. He has 32 years experience

as an engineer, a program manager, and an aerospace business unit leader. He also holds a BSEE in electrical engineering,

an MBA in business administration, an MA in human and organizational development, and a PhD in human and

organization systems. He is a member of the American Institute of Aeronautics and Astronautics, the Academy of

Management, and the Program Management Institute. He is also a certified Program Management Professional.

Galbraith, J. R. (1995). Designingorganizations. San Francisco: Josey-Bass.

Glaser, B. G., & Strauss, A. L.(1999, 1967). The discovery of groundedtheory: Strategies for qualitative research.New York: Aldine De Gruyter.

Graham, R. (1997). Managing theproject management process in aero-space and construction: A comparativeapproach. International Journal of

Project Management, 72(0), 39–45.Hall, R. H. (1962). Intraorganizational

structure variation. Administrative ScienceQuarterly (December, 1992), 295–308.

Hill, R. (1979). Matrix organizationand project management. University ofMichigan Press: 201979.

Johns, T. J. (1997). On creatingorganizational support for the projectmanagement method. International

Journal of Project Management, 62(8),47–53.

Knight, K. (1978). Matrix manage-ment. New York: Petrocelli Books.

Mintzberg, H. (1979). The structur-ing of organizations. Englewood Cliffs,NJ: Prentice-Hall.

Mintzberg, H. (1993). Structure infives. (2nd ed.). Englewood Cliffs, NJ:Prentice Hall.

PARADOXES OF CONTROL:THE (ELECTRONIC) MONITORINGAND REPORTING SYSTEM OF THEDUTCH HIGH SPEED ALLIANCE (HSA)

Recently, some the large public (trans-

port) infrastructures became an impor-

tant issue on the Dutch political agenda.

It was especially the High Speed Alliance

(HSA), one of the major transport infra-

structural projects in the Netherlands

that attracted a lot of attention. On one

hand, this project was highly advanced

from a technical point of view; on the

other hand, the HSA was characterized

by enormous (budgetary, time, and tech-

nical) problems. A large part of these

problems was supposed to be covered by

a complex (electronic) monitoring and

reporting system. The system, which was

established to exclude uncertainty and

risk, created its own uncertainty. In this

paper, the authors examine the role of

this system in the process of rationaliza-

tion and control within the HSA-organiza-

tion by focusing upon the process of

sensemaking. The authors argue that the

problems within the HSA can best be

understood in terms of a paradox: ration-

alization and control versus local, indi-

vidual freedom and initiatives. Three

major paradoxes have been distin-

guished: the cost paradox, the control

paradox, and the risk paradox.

Keywords: large infrastructural organiza-

tions; electronic project monitoring;

paradoxes of control


Vol. 38, No. 2, 75-83, ISSN 8756-9728/03

Introduction

During the last decade, public infrastructures became one of the main issues onthe Dutch political agenda, closely connected to discussions regarding eco-nomic development and growth, welfare, and mobility, especially within the

concept of the European community (Koppenjan, 2005). Among the many policyplans on infrastructures that were developed in different policy configurations—strategic (political) think tanks, the department of transport, the Dutch scientificadvisory board, etc.—two (initially interconnected) large-scale constructions proj-ects were defined in order to improve mobility and connect major economic coreslocated in different parts of surrounding European countries, in particular Germanyand Belgium: the Betuweroute project initiated in 1995, and the Dutch High SpeedAlliance (HSA) train (HogeSnelheidsLijn—HSL) project, which was started in 1996.

It was especially the HSA project that attracted a lot of (international) atten-tion, particularly because it was characterized by major deviations in terms ofoverspending, incorrect time frames/planning and corruption, which resulted inintensive parliamentary inquiries and a redesign of control models and steeringdesign. In 2004, the situation was so problematic that a Dutch parliamentariancommittee was established to investigate the cost and time planning problems ofthe HSA project.

About two years before the committee started its investigations, in2002/2003, the HSA organization was reorganized in a reaction to the internaltensions caused by technical and budget problems, and the external pressure bythe Dutch government to come up with transparent and reliable risk analysis(Koppenjan, 2005; Van Marrewijk & Veenswijk, 2005). The philosophy behindthe reorganization that was executed can be characterized in terms of New PublicManagement (NPM), which has been recently introduced, at least in rhetoricalterms, by policy-makers in order to improve the quality and transparency of gov-ernmental institutions (Pollitt, 2000). Referring to “the rise of the entrepreneur-ial state,” Osborne and Gaebler (1992) argued that in many western countries, a(neo-liberal) trend toward NPM leads to a “steer-not-row” division between(strategic) public policy areas and decentralized, separated implementation unitswith a strong use of market mechanisms.

This NPM philosophy has also been adopted in the Netherlands in manyways (Hendriks & Tops, 2003). For the HSA organization NPM meant a call forflexibility, transparency, and the execution of an unambiguous reporting system.

KEES BOERSMA, Vrije Universiteit AmsterdamSYTZE F. KINGMA, Vrije Universiteit AmsterdamMARCEL VEENSWIJK, Vrije Universiteit Amsterdam

ABSTRACT



However, is argued in the following, it was the HSA organi-zation that was established to oversee all kind of prob-lems—technical, budgetary, safety, political, etc.—thatbecame the major perpetrator of technical and budget prob-lems. It is this paradoxical situation that is addressed in thispaper by investigating how the (electronic) monitoring andreporting system of the HSA organization functions withinthe planning process. Additionally, the organizational cul-tural construction is examined and consequences of thesereporting systems of the HSA organization by looking athow relevant actors make sense of these systems is discussed(Weick, 1995). The sensemaking analysis will enable us tounderstand how individuals within HSA orient themselvesin situations that are complex and ambiguous.

The main research question in this paper is: How is the HSAproject internally organized, and in what way does the nature ofthe (electronic) monitoring and reporting systems influence therelations between the central and local actors? Also, how dothese actors make sense of the surveillance structure aroundthese contracting/control modes?

In the following sections, attention will be given to theorganizational and managerial background of the HSA as amega-project. Next, the outcome of the authors’ empiricalstudy into the (electronic) monitoring and reporting systemof the HSA will be presented. In this part of the paper, theuse of the systems as a sensemaking process will be lookedat. Arguably, tensions caused by the use of these monitoringsystems can be seen in view of a central organizational par-adox: rationalization and control versus local, individualfreedom. At the end of this paper, the authors will argue thatthis paradox can more generally be considered as character-istic for large-scale infrastructures.

Setting the Scene: The Managerial Culture of Mega-Projects

It has been argued before that cost estimates and control indecision-making and execution of transport infrastructuralprojects like the HSA are highly systematically and signifi-cantly misleading (Flyvbjerg, Holm, & Buhl, 2002). In 2003,Flyvbjerg, Holm, and Buhl presented their findings of a sta-tistical study of more than 250 infrastructural projects in 20nations. One of the most interesting outcomes of this surveywas that, at least in terms of costs, transport infrastructures donot perform as promised (Flyvbjerg, Holm, & Buhl, 2003).

Historically speaking, large-scale construction projectswere delivered under a Tayloristic control of just two parties(Clegg, Pitsis, Rura-Polley, & Marosszeky, 2002). The archi-tects were responsible for the design and the constructors forconstruction. With growing forms of complexity in large-scale projects, “the concept design and construct,” more andmore became the standard way of organizing this type ofproject, with government as a “third party.” In connection tothis, there is a worldwide transformation visible toward theso-called BOOT arrangements (build, own-operate-trans-fer), which has consequences for the way large-scale mega-projects are organized (see Dyer & Singh, 1998). The BOOTmodel requires a flexible organization to set up an out-sourcing infrastructure to provide and manage services for

the contracting parties. BOOT arrangements have beenintroduced as a promising structure in which a private sec-tor or organization cooperates with the government tobuild, own, or operate an infrastructure (or piece of) such asa railroad or an electricity plant.

However, these BOOT arrangements are typically high-risk ventures (Pestman, 2001). Such projects are not onlyextremely expensive, but also involve political controversies,complex organizational arrangements, and technologicalchallenges. They are long-term projects that are constantlychanging over time. The developments of the project or theoutcomes are therefore hard to predict and carry great uncer-tainties. The interesting point here is that in such technolo-gy-driven cases the organizational design anticipates thishigh-risk context and binds this project to periodic “progressreports,” in part based on specific “risk analysis” (Beck, 1992;Vaughan, 1999). Therefore, it can be seen in such situationsthat risk monitoring and reporting systems are often imple-mented as specific management tools not only designed foradministering the project but also intended for steering it.

The Cultural Aspects of (Electronic) Monitoring

and Reporting Systems

Interestingly, it was the Danish professor, Flyvbjerg, whowas consulted by the Dutch Parliament as the first leading,international expert on mega-projects on the committee oninfrastructure projects on 3 May 2004 to clarify his findings.The main problem the committee addressed was the misin-formation about Dutch mega-projects and their conse-quences. Possible solutions discussed by Flyvbjergconcerned (1) the Parliament rules and practices that had tobe changed in order to eliminate or reduce miscommunica-tion and (2) the HSA organizational accountability in orderto improve risk analysis (Flyvbjerg, 2004).

In fact, Flyvbjerg’s reaction and recommendations mustbe seen against the background of the HSA monitoring andreporting system. For many reasons, the management of theHSA organization introduced this system to meet both thecall for more transparency and integration. In line withclaims during the mid 1990s, the HSA organization imple-mented standardized business information systems. Withthese systems, organizations usually seek to improve produc-tion processes and services by enhancing management con-trol over complex processes, empowerment and motivationof personnel, increasing flexibility, efficiency, and effective-ness (Davenport, 1998; Frissen, 1996; Van der Ploeg, 2001).

Electronic management systems, such as adopted by theHSA, roughly concern two sets of information systems, com-puter-mediated communication systems (like intranet andknowledge management systems) and data management sys-tems (like enterprise resource planning, customer relation-ship management, and supply chain management). Thesesystems are often found together in complex organizationalsettings. They reflect the most recent phase in the informati-zation of organizations, integrating various business process-es within and among organizations. However, experience hastaught us that these systems, which are supposed to deliver


reliable (integrative) information, often do not live up tothese expectations. Standardized information systems do notonly offer business solutions, but also create a lot of neworganizational problems and unintended consequences(Boersma & Kingma, 2005). Instead of highly promisingtechnologies, standardized information systems are perhapsbetter understood as highly demanding technologies.

Problems associated with information systems are, atleast in part, attributed to an underestimation of the socialand cultural side of information systems. Organizational cul-ture will in this case less be operationalized as certain aspectsof organizations, and more in terms of a cultural perspectiveon organizations that considers culture as part and parcel ofthe entire organization (Alvesson, 2002; Czarniawska,1998). Technologies and organizations are from this per-spective studied as cultural phenomena in their own right;i.e., the meanings, norms, and values associated with partic-ular technologies and organizational practices (Orlikowski,2000). In particular, the concept of sensemaking (Weick,1995) is of analytical value here. In situations where peopleare confronted with rapid changes, complex problems andunexpected outcomes of new technologies, people literallymake sense of what is going on in the organization.

Research Strategy and Methods Used

In line with the idea of sensemaking, the authors’ methodwas to follow the qualitative analysis. The HSA case studytherefore, is based on document analysis, observations, andinterviews. It is the perceptions of interview respondentsthat gave the authors access to the ideas and conceptionsthat together form the webs of significance in and aroundthe organization (Czarniawska, 1998). The interviews wereopen and semi-structured in order to relate as closely as pos-sible to the everyday reality of the actors. For obvious rea-sons, no personal names are mentioned in this paper.

The authors analyzed the interviews in terms of (elec-tronic) management (including the reporting system), orga-nizational culture, the communication and negotiationprocesses, and power relations related to specific events ofthe HSA case. This includes asking probing questions aboutelements that the respondents consider to be important. Theauthors could meet the demands of analyzing the differenti-ation, and even fragmentation, of meanings across sociallydistributed activity systems that provide “technical supportfor their ongoing, local negotiation” (Martin, 2002).

In addition, the authors got access to relevant companydocuments to reconstruct the dynamics of the HSA project, inrelation to the monitoring and reporting system. The docu-ments studied in this research were quite diverse. There were“official” documents and policy papers, but also documentswritten for one’s own use, like memos, accompanying letters,(concept) notes, and minutes. Some of these documents arepublicly available on the Internet websites of the HSA organ-ization (see http://www.hslzuid.nl/hsl/uk/hslzuid/). Eventhough most of the studied documents came from the organ-ization itself, documents written about the organization werealso looked at. Because the HSA project engendered many

public discussions and debates in the Netherlands, publica-tions in newspapers and other attention in the media werealso studied. As part of this, a media analysis was a relevantpart of the analysis of documents in the case.

The Monitoring and Reporting System of the HSA:

A Sensemaking Process

The basic concept of the HSA consisted of a plan to connectDutch backbone, Amsterdam, by means of a 100-km sepa-rate railroad toward the Belgian border. The HSA project wasdefined as a result of a political discussion on future mobil-ity in the Netherlands in relation to surrounding countriesin 1996 (see Van Marrewijk & Veenswijk, 2005). The projectdesign that followed was “new” in the sense that the projectgroup—directed by a young entrepreneurial project direc-tor—rejected a “sound blue print” plan in advance (despitegovernmental pressure), but preferred a participative modelof project development.

The project was originally designed as an experimentalpublic-private constellation in which public and privateactors participated and had a joint responsibility for con-struction and design, as well as for exploitation. Most of thedesigning was actually the result of the public-private(BOOT) arrangement, and this hybrid organizational con-struction was reflected in the way the project organizationswere empowered (70% of the employees were hired on atemporal basis, 30% were public employees), exploitationremained a matter of the state. This also counts for the (pub-lic sector) hierarchical—machine like—formal construction,which was chosen in order to build the project. The topmanagers of the HSA project organization were all publicemployees. The CEO (in Dutch called Hoofdingenieur-directeur abbreviated as the HID—the engineer who isresponsible for the technical part of the entire project)reported by the secretary-general of the Ministry ofTransport to the minister. In 2003, the HSA project wasadministratively separated from other infrastructural proj-ects like the Betuweroute and further implemented as a sep-arate mega-project.

The project was divided in five geographic entities: everyentity contained a set of unique infrastructural artifacts likebridges, tunnels, etc. All five geographic entities were inter-connected via loosely coupled separate project agencies,which all reported to a principal organization (called theCentrale Projectorganisatie) in Zoetermeer, a middle-sizedcity in the Province of South Holland. The interconnectionwas realized through various ranges of contracts, between(1) the principal and the agencies and (2) between privatearchitects, or constructors, and the agencies. In total, therewere more than 900 (sub)contracts in operation (Ministerievan Verkeer en Waterstaat, 2003, 2004). These contractswere monitored on a quarterly basis, and surveillance tookplace using a complicated set of ICT systems.

First, the reports were based upon an integrated busi-ness system (an ERP system). However, from an ICT point ofview this system is not highly sophisticated, but rather frag-mented. The content of the system is the outcome of a huge


amount of Word documents, e-mails, and Excel spread-sheets that are collected by the (sub)contract managers andthat summarize technical and organizational informationabout subprojects. Second, the quarterly reports contain sev-eral “standard issues” (for instance, in terms of time andcost: technical development, risk calculation, human capac-ity, environmental developments, and communication) forcentral management, for which the information must becollected. This means that the monitoring and reporting sys-tem is standardized by project features, but not with thehelp of a standardized software package readily available onthe market.

A typical quarterly report looks as follows. Each quar-terly report starts with a detailed management summary inwhich the main problems are described. The body of thereport contains three important fields in the risk analyses—which is made visible by a Monte Carlo risk simulation:time, technology, and money. External effects, such as pub-lic safety issues, and internal processes, like communica-tions, are covered in a separate section. The report ends withan average of eight appendices in which completed tablesare included. The tables consist of information based uponnumerous subprojects’ electronic reporting systems.

Project managers expressed that it is especially the man-agement summary that is the topic for an ongoing discus-sion, because it is this information that is used for theannual reports of the entire organization. That means that itis in the summary that the data is interpreted and translatedby the responsible (project) manager. For the project man-agers, the report has basically two functions. First, it is seenas a checklist used as an afterward justification of choicesmade in the planning process. In this process of justifica-tion, the most important standardized issues are checkedoff. Second, the report is used as a back-up system: on thebasis of the data mentioned in the reports, a manager isalways covered in the time of technical and budgetary prob-lems. In this way, it is used to protect oneself against future,possible problems.

Central Versus Local: Tensions in the Organization

During the start of the HSA project in 1996/1997, the basicresponsibility regarding the accuracy and planning of thedata in the reports was with the contractors. The organiza-tional philosophy was that responsibility should be “decen-tral unless…,” which meant that central interventions wereonly allowed in specific situations—in other words, theremust be a good reason to intervene (Van Marrewijk &Veenswijk, 2005). In doing so, the HSA organization fol-lowed a well-known and traditional principle, since decen-tralization of operating activities has been associated withefficiency advantages caused by improved information man-agement (the use of local knowledge) and managementopportunism (by less informed central managers) (e.g.,Williamson, 1985). That means that on the central level ofthe organization, the bureaucratic rules are leading; on thelocal level, however, there is ample room for entrepreneur-ship and locals initiatives (Courpasson, 2000).

More and more action plans regarding the “local set-ting” were designed within the subgroups, headed andlegitimized by professional “project-agency managers,”which by the beginning of 2002 caused major conflictsbetween the principal (specifically the earlier-mentionedHID) and agencies or project-agency managers. Centralissues in conflict were: report-structure and intervals, lackof trust on accuracy of data between different parties, andinternal communication structure (who does what).Almost all of the issues can be related to the central andgeneral aspects of image/identity creation: “principalwants to control us and rule everything out via proce-dures” versus “agencies don’t want to share data and doeverything in their own (probably malicious) ways.”

These problems can be seen as a negative side effectof decentralization, because this strategy is less effectiveat generating knowledge and innovation that transcendsthe level of the business unit. In a highly decentralizedorganization, it is hard to control heterogeneous infor-mation collected and reported by the individual busi-ness units. It is this problem that caused for tensionswithin the HSA organization. The organizational con-flict led to a set of interventions backed by the DutchMinister of Transport.

The most important and visible interventions in theorganization were, first, the assignment of a new topexecutive who was held responsible for the whole risk-management process and who became the mouthpieceof the organization. Second, to meet the problem offragmentation, the organizational philosophy wasadapted to “central unless.” And third, a set of organiza-tional development measures were taken, includingteam building, dialogue days between different groups,restructuring of contracting relations, and a job rotationof managers who were not supporting the new course ofaction. In addition to these matters, new contractingprocedures were introduced and a dramatic clustering ofcontracts took place.

What is striking here is that the (electronic) moni-toring and reporting systems that were introduced, aspart of a process of rationalization (unintentionally)resulted in contradictions within the organization. Suchcontradictions, which can be understood in terms of acentral paradox—(1) rationalization and control and (2)freedom—cannot be found when complex organizationsare studied (see Koot, Sabelis, & Ybema, 1996).

Beyond the Borders of Control: A Triple Paradox

In this section, the triple paradox will be used to reflectupon the HSA case as previously described. To be moreprecise in the analysis, three major paradoxes have beendistinguished: the cost paradox, the control paradox,and the risk paradox. These paradoxes will be describedand each paradox will be illustrated by an empiricalexample. With the help of the paradoxes, the authorswill show how, in concrete situations, the HSA actorswere confronted with complex issues of control.


Cost Paradox: Poor Performance Versus New Project InitiativesAccording to Flyvbjerg, Holm, and Buhl (2002) the cost par-adox in large-scale infraprojects consists in the irony thatmore and more mega-projects are constructed and semi-autonomous public implementation routes are realized,despite the poor performance record of many of these proj-ects. This image of a mega-project suggests that these orga-nizational forms should be or could be rational systemsorganized to operate as efficiently as possible (Klijn, 1997).Such projects include a division based on the assumptionthat complexity is located in the domain of policy formula-tion (development and design) and that project organiza-tions (construct) are situated in a social environment inwhich work-related goals and activities are clear and can be(fully) predicted, political backing is stable, there are noresisting single issue groups, and no internal conflicts thatlame the project.

The cost paradox manifests itself in different ways in theHSA case. At several times during the implementation phasethe managers of the central HSA organization were con-fronted with local problems that were hardly predictableand were caused by the organizational distance betweendevelopment and design, and local constraints. A strikingillustration is formed by the problems that arose about fouryears ago with regard to the height of gates at the stations sit-uated along the HSA railway. Due to a series of Europeandirectives by the European Transport Committee—e.g., theDirective 2001/16, which is part of the EU Safety Rules—theDutch national rail transport organizations were forced tostandardize the high-speed rail system. These directives areproclaimed by the EU to ensure the interoperability of thenetworks, not only in the field of technical standardization,but also in the field of passenger safety. One of these direc-tives directly concerns the height of stages and it was thisdirective that caused tensions in the HSA organization.

The problem was that this directive was implemented ata time when the technical blueprints of the stages werealready completed. On these technical blueprints, the stageswere designed much lower than the standardized height laiddown in the EU directive. Of course, redesigning the stageswould be very expensive. Strikingly, on a local level it waswell known that during the designing process the directivewas in preparation, as one HSA manager told the authors:“… at that time we knew that something was going to hap-pen, but they [the HSA direction, KB, SK, MV] did not knowthat. Every now and then, however, we take the time to reada EU directive …” However, the problem caused by the EUdirective was not included in the quarterly reports.

This situation forced the actors involved to come upwith (ad-hoc) solutions. A possible solution put forward bythe central organization was to adapt the boarding-floor ofthe trains, but of course this solution would be even moreexpensive and would involve higher costs by the suppliers.Eventually, the only solution was to redesign the stages andto make new blueprints.

Understandably, this situation not only made the blue-prints much more expensive, but it also gave rise to frustra-

tions both at the central and at the local level. It becameobvious that the problem was brought about by the fact thatthe HSA project was organized as a mega-project. The para-dox here is that the central HSA organization, which wasestablished to control the budget of the Dutch high-speedrail system, led to the central actors (i.e., the HSA direction)wearing blinkers, as a consequence of which local problemswere overlooked. The local actors were not always able toconvince the HSA direction that local constraints can havean impact on the organization as a whole.

Control Paradox: Sophisticated Control Mechanisms VersusUnexpected Control GapsThe control paradox can be found in the situation inwhich the characteristics of the organization, which arebrought about in order to exclude uncertainty, easily canlead to a situation in which knowledge exchange amongindividuals is restricted (by the formal rules of the organ-ization). A version of this paradox is also mentioned byVaughan (1999, p. 914) who states that the organization-al structure of large, complex organizations leads to “… acentral paradox in organizations: namely, that the char-acteristics usually associated with the bright side oforganizations—the structures and processes designed toassure certainty, order knowledge, and stabilize opera-tions, thereby making coordinated activity possible—alsohave their dark side—the capacity to generate uncertain-ty, disordered knowledge, instability and unanticipatedoutcomes.”

Vaughan’s work dealing with the complexity of taskrelation draws attention to the multiple identificationand task interpretation within semiautonomous workinggroups. Her detailed analysis of the Challenger, the NASAspace shuttle that exploded, reflects subcultural fragmen-tation and cultural trading zones in which “reality isnegotiated” with ambiguity as the leading concept. It notonly pinpoints the rationale of bureaucratic actors to cre-ate a habit of normalizing deviations from safety proce-dures, but also underlines the interlocked behaviorsbetween the actors in the working groups, as well as theirorientation to the technical artifacts that were construct-ed to reproduce the safety regulations, as also noted byWeick (1995).

When designing the HSA trajectory, much thoughtwas given to monitoring and eliminating the so-called“external effects.” We are talking about calculated, unwel-come situations, which could come up in the area of spa-tial aesthetics and the environment. These effects weredefined and categorized as much as possible in advanceby means of so-called impact assessments.

In the HSA quarterly reports, an account is given ofthe degree to which these effects fall within acceptablesocial, economical, and political preconditions. In addi-tion, they report on whether supplementary measures arecalled for. Despite this principle of calculated control, aparadoxical situation has come up more than once duringthe day-to-day affairs of the project. An example of the


control paradox in the HSA project is the matter of theinordinately high mortality rate among birds along thetrajectory North Brabant (a Dutch province), whichoccurred toward the end of 2003.

To limit the effects of noise pollution as much as possi-ble, transparent noise barriers were erected along the trajec-tory in North Brabant on a grand scale in 2002. The noisebarriers had been the subject of some political controversybecause of the high attendant costs and the drastic effects onthe landscape. Many political parties thus had a strong focuson the actual noise reduction, which the erection of the bar-riers would yield, as well as the aesthetic valuation of theseemingly endless barriers. The green parties, in particular,made an issue of the effects they would have for people liv-ing in the immediate vicinity of the barriers.

The noise barriers appeared to be a success. Indicativenoise measurements taken in 2002 revealed that noise levelshad dropped by 70%, and people living in the vicinity of thebarriers were generally positive about the manner in whichthey had been integrated into the landscape. On the basis ofthe quarterly reports, the political parties debated with fer-vor on this issue. Green parties claimed the successful atten-tion for environmental issues in the technical design, whileother political parties underlined the importance of thepublic interest, which they had brought up.

Late 2003, it became known that an excessive numberof dead birds had been found along the much-praised noisebarriers. Counts by bird protectionists led to a tally of morethan 5,000 dead birds, including a number of unusualspecies. There was little attention paid to this issue in theHSA reports at first, but after an alarming letter toParliament from Vogelbescherming Nederland, the Dutchpartner of Birdlife International, the Minister for Transportordered an investigation. The outcome was that the highmortality of birds was the result of a one-sided fixation onnoise and design and an inadequate appraisal by the man-agement of the project of other possible effects. To readdressthe situation, a proposal was made to affix vertical stripes tothe barriers as quickly as possible. The additional costs forthis were estimated at 11 to 18 million euro (source: the[Dutch] Ministry of Transport, Public Works and WaterManagement, News Log, July, 2004).

In conclusion, we can deduce from this example thatthe essence of the control paradox lies in the fact that theelimination of uncertainties by means of the calculation of“planned external effects” as in the case of the HSA projectautomatically brings forth its own uncertainties.

Risk Paradox: Risk Analyses Leading to Illusions of Controland Risk TakingThe risk paradox describes a situation in which the adop-tion and use of risk analyses as the dominant managementtool led to an increase in risk taking in decision-makingprocesses, instead of the intended decrease in risk taking orthe “taming” of risks. The taming of risks refers to therational, calculative aspect of risks in which problems aredefined as mechanical, consisting of parts that when bro-

ken down attract fixed, linear, optimal solutions (Holt,2004). Such an adoption of risk analyses may involve orga-nizational cultural shifts that are characterized by a reflex-ive attitude toward the unintended consequences and sideeffects of modern organizations.

Beck (1992) coined this shift as “reflexive moderniza-tion,” when we are being confronted with these conse-quences and effects, recognizing and anticipating them,reacting upon them and more or less accepting the poten-tial consequences as well as making organizationsaccountable. The term “reflexive” refers not only to a moreor less direct reaction to the consequences of former deci-sions but also to “reflection” in a sense of awareness. Risksin this context do not refer to actual harm or to theattempts to remove danger, which is the object of risk;rather it is the reshaping of organizations as a reaction tothe possibility of (often unknown) dangers. Ultimately, ifrisks are being acknowledged, this may change the condi-tion of their existence (Van Loon, 2002).

In 2000 the HSA organization was remodeled aftersuch a risk model of management. The adoption of thismodel signaled a new phase in the project organizationcharacterized by a shift from decentralization and productorientation toward centralization and a process orienta-tion. This became the dominant mode of management.New management procedures were introduced, notablythe system of quarterly reports based on risk assessmentsby contract managers. Within the new system, contractmanagers became the central actors, translating many con-struction issues within the monitoring and reporting sys-tem in quantifiable, preferably financial terms, on whichmanagement decisions could be based.

The unexpected problems in the construction of thelarge, two-kilometers bridge crossing the water of theDutch river “Hollandsch Diep,” may serve as an exampleof how this new system of risk analyses worked and alsoof the paradox involved. The construction problem sur-faced in October 2004 when one of the 11 pillarsappeared to subside. The pillars were only allowed tosubside 50 millimeters per century, but this particularpillar had subsided more than five centimeters in onlytwo years. Immediately after the discovery of this subsi-dence it was defined as an enormous risk to the HSAproject. In the worst case scenario, the pillar had to becompletely reconstructed, which would cause seriousdelays and enormous costs for the entire organization. Asimple test was designed in which the contractor parked16 trucks with a total amount of 750 tons of sand on thepillar, to see to which extent the pillar would furthersubside. Fortunately, management was confident withthe test results and decided to continue the constructionof the bridge. However, there was serious debate con-cerning this incident about a possible cover up of possi-ble calculation errors by engineers involved in theconstruction of the bridge. In February 2005 the Ministerof Infrastructure declared, after a careful investigation,that this was not the case.


This case illustrates how the advance in the constructionof the HSA is monitored and managed in terms of riskassessments. It also makes clear that the assessment of risk isnot only a technical issue but also an organizational andsocio-political issue. The ultimate paradox in this case is thatthe risk assessment, the test with the trucks, served to ration-alize and legitimate the continuation of the infrastructuralconstruction works in spite of the fact that there seemed tobe something wrong with the pillar.

Discussion: Lessons Learned

What can we learn from the HSA case? From the start of HSAproject, much store was set on innovation by means of adecentralization-oriented strategy. The (one-sided) focus oncontracting was seen as a crucial vehicle in the battle againstoverspending, but this strategy appeared to be insufficient.The case brings us to three—heuristic—lessons (see also VanMarrewijk & Veenswijk, 2006).

Lesson one: “Know who is responsible.” The separation strat-egy of HSA leads to structural unproductive power strugglesbetween the innovative core and the “mother” organization.Although “in theory” the project responsibilities were

clear, the HSA project team became ensnared in an organi-zational spider’s web of parties, which all claimed the con-trol of the project. On the front stage, it was the Ministerand the top management of the Ministry of Transport whodirected the project and were responsible for the qualityand dependability of the project. In actual practice, howev-er, the project team and the ministry reached a state of(apparently) continuous conflict about concluded contrac-tual agreements, costs, and responsibilities. The phasedtransfer of responsibilities appeared to be a planned, con-trollable exercise but, because of the impossibility of set-tling on a clear delineation of tasks, turned out to becompletely unpredictable.

Lesson two: “Contracting does not solve the issue of respon-sibility.” In HSA, the one-sided fixation on “partial” con-tract indicators becomes an important barrier for integralproject evaluation.In the HSA case, the quarterly contract reports more and

more acted as internal benchmarks and vehicles for adminis-trative and organizational improvements for central manage-ment. Although this system suggests a clear-cut vision onproject evolvement, in actual practice there was no integral sys-tem in which progress of the interrelated parts was monitored.The (relatively) separated contracts were related to specific arti-facts (e.g., bridges, tunnels, platforms) and cost overrun onthese subcontracts did not act as a trigger for redesign of theother related project artifacts. In fact, contract realty in HSAbecame more and more detached from the day-to-day organi-zational practices and the problems that were experienced interms of internal cooperation, communication, team spirit,and trust in the quality of the project outcome.

Lesson three: “Separation leads to (a certain amount of)professional ‘ivory tower’ entrapment.” In HSA the cre-ative entrepreneurial project team isolates and marginal-izes themselves.

In the different stages of the HSA project thereappeared to be large gaps between the creative projectexperts and the “bureaucratic” managers located at theMinistry of Transport. The initial project team became high-ly isolated as their innovative, participative model of projectdevelopment turned out to be a large financial risk. The pro-fessionals of the project team were convinced of their own“success formula” and focused themselves on the positiveresponses from their project environment. Critique wasdenied or seen as an act of a “power play” of managementof the Ministry of Transport. A set of heavy organizationalinterventions, earlier mentioned, such as the assignment ofa new top executive, the shift to a new ‘central unless’ phi-losophy, the start of so called ”dialogue days” betweendecentral project professionals and central executives brutal-ly interfered with this view and dramatically change theproject culture.

In order to tackle the interface problems that occurs inthe different project stages, some large-scale projects haveintroduced the concept of the “cultural broker”; an actorwho is able to bridge the gap by negotiating between dif-ferent project realities. In the case of the HSA, such anactor might have been able to prevent the situation of proj-ect entrapment.

Conclusion: Paradoxes of Control at the Dutch HSA

In the case of the HSA, the cost factor became the majorstruggle for management as well as for political representa-tives. As shown, the original decentralization-oriented strat-egy pushed toward a large variety of types and amounts ofinternal contracts. Although the one-sided focus on con-tracting was seen as a crucial vehicle in the battle against costoverrun, this strategy was insufficient. The (relatively) sepa-rated contracts were related to specific artifacts (bridges, tun-nels, platforms, etc.) and the cost overrun on thesesubcontracts did not act as trigger for redesign on the otherproject artifacts. On the contrary, local actors often consid-ered the completion of an artifact like a tunnel (despite thecost factor) as an act of strong management and decisive-ness, a result of courage to “stand up” against centralbureaucrats that should be followed by the other projectagency managers.

In the last phase of the project, the (new) project directorof the HSA organization heavily increased contract standardi-zation. Project controllers were empowered to intervene ifnecessary. The quarterly reports were based on the complicat-ed reporting procedures and ICT systems. The reports aboutthe various subprojects more and more acted as internalbenchmarks and as vehicles for administrative and organiza-tional improvements (“standard issues,” for instance, were—in terms of time and cost—technical development, riskcalculation, human capacity, environmental developments,and communication) for the central management. At thesame time, contract reality became detached from the day-to-day organizational practices and the problems that were expe-riences in terms of internal cooperation, communication,team spirit, and trust in the quality of the project outcome.


Although the (quarterly and annual) reports wereavailable from the intranet, only a small group of profes-sionals were able to read and interpret the report results.Project-agency managers “translated” the results in termsof their own organizational reality and strategically pre-sented the information for their own purposes. This is tobe seen as an important part in the process of sensemak-ing—after all, in this way actors created meanings abouttheir social setting in interaction with others (Weick,1995). Due to these “translation” activities, a serious gapevolved between (1) the “front stage” realities of the proj-ect dominated with a mixture of rational models, figures,and schemes of the project and political definitions aboutfailures, and (2) the “backstage” day-to-day reality of theorganizational group.

The political inquiry started not only as a ritual (andact) of political accountability, but also as a search for“project improvement,” especially in relation to the themeof changing environmental conditions only partially metthe intended goals. Entrepreneurship, innovation, andcourage, marked as critical issues by almost all of key play-ers were disqualified by cost-related arguments. In the caseof the HSA, the flexibility paradox was almost personifiedvia the demotion of the former project director. The samepolitical actors that initially supported him in his unortho-dox—flexible—way to organize and build the project nowdismissed him for the lack of (contractual) predictabilityand procedural correctness.

References

Alvesson, M. (2002). Understanding organizational cul-ture. London: Sage.

Beck, U. (1992). Risk society: Towards a new modernity.London: Sage.

Boersma, F. K., & Kingma S. F. (2005). From means toends: The transformation of ERP in a manufacturing com-pany. Journal of Strategic Information Systems, 14, 197–219.

Clegg, S., Pitsis, T. S., Rura-Polley, T., & Marosszeky, M.(2002). Governmentally matters: Designing an alliance cul-ture of interorganizational collaboration for managing proj-ects. Organization Studies, 23(3), 317–337.

Courpasson, D. (2000). Managerial strategies of domi-nation. Power in soft bureaucracies. Organization Studies,21(1), 141–159.

Czarniawska, B. (1998). A narrative approach to organiza-tion studies. London: Sage.

Davenport, T. H. (1998). Putting the enterprise into theenterprise system. Harvard Business Review, 121–131.

Dyer, J. H., & Singh, H. (1998). The relational view:Cooperative strategy and sources of interorganizationalcompetitive advantage. Academy of Management Review,23(4), 660–679.

Flyvbjerg, B., Holm, M. K. S., & Buhl, S. L. (2002).Underestimating costs in public works projects. Error or lie?American Planning Association Journal, 68(3), 279–295.

Flyvbjerg, B., Holm, M. K. S., & Buhl, S. L. (2003). Howcommon and how large are cost overruns in transport infra-structure project? Transport Reviews, 23(1), 71–88.

Flyvbjerg, B. (2004). Improving the role of Parliamentin decision making and realisation with regard to infra-structure project. Lecture held in The Hague in theCommittee on Infrastructure Projects, May 3, 2004.

Frissen, P. H. A. (1996). De Virtuele Staat. Politiek, bestu-ur, technologie: een postmodern verhaal. Schoonhoven:Academic Service.

Hendriks, F., & Tops, P. (2003). Local public manage-ment reforms in the Netherlands: Fads, fashions and windsof change. Public Administration, 81(2), 301–323.

Holt, R. (2004). Risk management: The talking cure.Organization, 11(2), 251–270.

Klijn, E. H. (1997). Policy network, an overview. InKickert, W., E. H. Klijn, & J. Koppenjan (Eds.), Managingcomplex networks (pp. 14–34). London: Sage.

Koot, W., Sabelis, I., & Ybema, S. (Eds.). (1996).Contradictions in context. Puzzling over paradoxes in contempo-rary organizations. Amsterdam: VU University Press.

Koppenjan, J. F. M. (2005). The formation of public-pri-vate partnerships: Lessons from nine transport infrastructureprojects in the Netherlands. Public Administration, 83(1),135–157.

Martin, J. (2002). Organizational culture. Mapping the ter-rain. London: Sage.

Ministerie van Verkeer en Waterstaat. (2003).Voortgangsrapport 13 Hogesnelheidslijn-Zuid, Den Haag.

Ministerie van Verkeer en Waterstaat. (2004).Voortgangsrapport 14 Hogesnelheidslijn-Zuid, Den Haag.

Orlikowski, W. J. (2000). Using technology and consti-tuting structures: A practice lens for studying technology inorganizations. Organization Science, 11(4), 404–428.

Osborne, D., & Gaebler, T. (1992). Reinventing govern-ment. How the entrepreneurial spirit is transforming the publicsector. New York: Addison-Wesley.

Pestman, P. (2001). In het spoor van de Betuweroute.Mobilisatie, besluitvorming en institutionalisering rond een grootinfrastructureel project. Amsterdam: Rozenberg Publishers.

Pollitt, C. (2000). Is the emperor in his underwear? Ananalysis of the impacts of public management reform. PublicManagement, 2(2), 181–199.

Van der Ploeg, R. (2001). De invloed van ICT opmaatschappij en overheid. Naïef vooruitgangsgeloof of hardewerkelijkheid? Amsterdam: AUP/Salome.

Van Loon, J. (2002). Risk and technological culture:Towards a sociology of virulence. London: Routledge.

Van Marrewijk, A., & Veenswijk, M. (2005).Anthropologists at work in the tunnel: Cultural interventionand change in the HSL-Zuid project. In Van Marrewijk, A. &Verweel, P. (Eds.), Exploring organizations. The development oforganizational anthropology in the Netherlands (pp. 45–55).Amsterdam: SWP Publishers.


Van Marrewijk, A. H., & Veenswijk, M. (2006). The cul-ture of project management. New York, London, Amsterdam:FT/Prentice Hall.

Vaughan, D. (1999). The role of the organization in theproduction of techno-scientific knowledge. Social Studies of

Science, 29(6), 913–943.Weick, K. E. (1995). Sensemaking in organizations.

London: Sage.Williamson, O. E. (1985). The economic institutions of

capitalism. New York: Free Press.

KEES BOERSMA is associate professor at the Vrije Universiteit Amsterdam in the Department of Culture, Organization and

Management. His research interests are in science and technology studies, business history and history of technology,

and organization, culture and power. He published widely on R&D history, organizational learning, and organizational

culture. His publications appeared in such journals as Enterprise and Society, History and Technology, Human Relations,and Journal of Strategic Information Systems. He teaches in the courses of organizational behavior, organizational politics

and technology and culture.

SYTZE F. KINGMA is lecturer in the field of organizational space and technology at the Department of Culture, Organization

and Management, Vrije Universiteit Amsterdam, NL. His research interests involve the confrontation between the

“material” and the “virtual” dimensions of organizational networks, and the way “risk” is implicated in contemporary

organizational contexts. He published a book (in Dutch) titled The Gambling Complex (2002). His major international

articles appeared in International Gambling Studies, Journal for Strategic Information Systems, and Culture andOrganization.

MARCEL VEENWIJK is full professor in management of cultural change at the Department of Culture, Organization and

Management of the Vrije Universiteit Amsterdam. He publishes on organizational culture, change processes, and

intervention strategies, especially in the context of public sector organizations and has a wide experience as researcher

as consultant. He is editor-in-chief of Intervention Research: International Journal on Culture, Organization andManagement. In 2006 he co-published The Culture of Project Management. Understanding Daily Life in ComplexMegaprojects.


CONTAINING TRANSACTION COSTS IN ERPIMPLEMENTATION THROUGH IDENTIFICATIONOF STRATEGIC LEARNING PROJECTS

Implementation of enterprise resource plan-

ning (ERP) systems requires in-depth under-

standing of the idiosyncrasies of the receiving

organization, thus implying specific learning

costs. ERP projects delivered in form of con-

tracted projects in a competitive market envi-

ronment often ignore costs for competitive

reasons (Linderoth & Lundqvist, 2004).

However, these learning costs add to the over-

all transaction costs of a project. If not bud-

geted into the individual projects, these costs

accumulate at the project portfolio level of

project delivery organizations (sellers) and

must be managed adequately. One way of

minimizing learning costs at the portfolio level

is to identify strategic learning projects, which

accrue higher than planned (competitive)

costs, but lower the costs of future projects

and thereby the overall learning costs at the

portfolio level. The present study investigated

the practices of dealing with costs of learning

projects at a Swedish ERP consultancy over a

period of six months. The results show how to

lower portfolio-level project costs in the long-

term through identification of strategic learn-

ing projects, with expected budget overruns

in the short-term. The study takes a transac-

tion-costs-economic (TCE) perspective. A

model shows the impact of short-term learn-

ing investments on the lowering of implemen-

tation risks in future projects. The model can

be used by managers for understanding long-

term profitability by exceeding time and budg-

et objectives in the short term.

Keywords: projects; knowledge creation;

transaction cost economics; ERP; multi-

project organization; portfolio management


Introduction

An increasing need for customer-specific solutions made the project form of organ-ization the structure of choice for delivery of technology and services to businessclients (Turner & Keegan, 1999). Projects are typically defined as follows:

A temporary organization to which resources are assigned to undertake a unique,novel and transient endeavour managing the inherent uncertainty and need for integra-tion in order to deliver beneficial objectives of change (Turner & Müller, 2003, p. 7).

The type of projects focused on in the present study are enterprise resourcesystems (ERPs), which automate and integrate the different functions in a firm,like accounting, manufacturing sales in one organization-wide system of interre-lated applications. ERP systems are typically bought from a vendor that offerscompetitive prices in the open market. Implementation of IT systems in general,and ERP system in particular, require in-depth understanding of the buying orga-nization’s processes and tasks in order to achieve the levels of automation requiredto accomplish the projects’ business objectives. The seller organization’s learningabout the idiosyncrasies of the buyer organization is often accomplished in jointefforts involving resources from buyer and seller organizations. Given the com-petitive nature of the bidding process, selling organizations have to bid at the low-est possible prices and therefore often underestimate or even ignore the requiredlearning of the seller organization’s consultants about the buying firm’s organiza-tion. Consequently, projects do not meet targeted budget and time objectives. Plandeviations often occur early in the project, caused by incorrectly budgeted learn-ing efforts. Because of these deviations, the in-depth learning efforts are kept at theabsolute minimum, otherwise the project risks becoming a failure if it exceedstime and budget objectives (PMI, 2004). This classical interpretation of projectsuccess as achievement of planned time, cost, and quality objectives still prevails,as shown by Jugdev and Müller (2005). It emphasizes application of traditionalrisk minimizing techniques in form of contracting, planning, and control to pre-vent exceeding budget and time frames agreed upon between the buyer and thesupplier organizations (Bajaj, 1997; Jiang, 2002; Kumar, 2002; Lind, 2002; Mills,2001; Pattersson and Neailey, 2002; Turner, 1999).

Given that ERP projects are mostly implemented as a number of projects overtime, to limit learning at the outset of these projects might not be the best solu-tion in the long term. Success should be measured on the totality of costs of all

ANNIKA ANDERSSON, Umeå School of Business, SwedenRALF MÜLLER, Umeå School of Business, Sweden

ABSTRACT


projects with a client. This is achieved by allowing the appli-cation of appropriate opportunities for learning about thereceiving organization early on in the first project, even ifthis causes overspending on budgets and time frames of thefirst project(s). This poses a risk to the supplier, but also achance to minimize costs in the long term. The level of risktaking is influenced by the risk tolerance of the organiza-tion, the project manager, and the stakeholders (Kwak &LaPlace, 2005). Turner and Keegan (1999) followed similarlines of thought and showed the need for freedom to dodevelopments not planned for originally in a project.

Defining ERP projects as learning platforms could helpin setting realistic expectations in terms of probability formeeting project objectives. Within this study these ERPlearning projects are seen as a platform for knowledge devel-opment. A learning ERP project contains a considerable, butnot necessarily paid for, part of knowledge development toenable appropriate implementation of the system into thereceiving organization. However, stressing the need forlearning in projects is often viewed as risky by the stake-holders, because those projects often fail to reach time andbudget objectives. Such a project is therefore to be evaluatedon both the learning results and the project results.

If not accounted for in the individual projects, the costoverruns just described accrue at the portfolio level of thesupplier organization. A portfolio is defined as “a collectionof projects [...] and or programs (a group of related projectsmanaged in a coordinated way to obtain benefits and con-trol not available from managing them individually) andother work that are grouped together to facilitate the effec-tive management of that work to meet strategic businessobjectives” (PMI, 2006, p. 4).

The aim of this study is to identify how knowledgedevelopment in ERP projects can be used to minimize thelong-term transaction costs over several projects that an ERPsupplier organization has with a client. Transaction costs arethe totality of costs arising from delivery of a system to aclient, thus comprise the following costs in addition to thecontracted project costs (Turner & Müller, 2004a):• Identification, accreditation, and selection of suppliers• Performance monitoring• Configuration management• Processing of variations• Performance feedback.

The research question addressed within this study is: Howcan learning costs in ERP projects be minimized in the long term?

The unit of analysis in this study is the selling (delivery)organization. The study assesses the long-term, learning-related transaction costs, from the perspective of a supplierorganization delivering several ERP projects to one client.

Contracted projects, defined as learning projects, is not atopic often followed in project management research. Thereis a lack of research combining knowledge creation, TCE per-spective, and a short- and long-term perspective. Of interestis the supplier’s perspective on learning projects, becausethey usually implement these projects in accordance with

existing standards for success evaluation, such as meetingtime, cost, and quality goals. Moreover, the supplier oftenworks in a multi-project setting with other projects operatingin parallel, while a series of sequential projects with a clientis implemented. Thus, it is interesting for both academia andindustry to find out how a learning perspective could affecttransaction costs both in the short- and the long-term.

The paper continues with a summary of the relevant lit-erature, followed by the research propositions.Subsequently, the method and data collection is described,followed by the analysis results and a model that shows theimplications of knowledge development in a multi-projectrelationship with one customer. The paper concludes withthe managerial and theoretical implications of the results.

Literature Review

The literature reviewed addresses the following:• Transaction costs economics• The control perspective within projects• Learning projects.

Transaction Costs EconomicsTCE focuses on the level of transactions that converts“input” to desired “output,” for example, the projects com-prising the implementation of a new ERP system to improvea company’s internal efficiency. According to Williamson(1996), transaction costs are the comparative costs of plan-ning, adapting, and monitoring of the completion of atransaction under alternative governance structures. TCEproposes that firms adapt their governance structures toachieve the lowest possible transaction costs. This is typical-ly done by choosing the most appropriate contract typebetween buyer and seller. Williamson (1975) identifiedthree dimensions of TCE:• The degree of asset specificity as the main influential fac-

tor. This is the extent to which the object of the transactionis specific (or unique) to the individual transaction andcannot be redeployed in future transactions.

• The degree of uncertainty that arises (a) from a lack ofcommunication or conscious supply of false and mislead-ing signals, preventing a decision-maker finding out aboutthe plans made by others in the transaction; and (b) thegeneral uncertainty in human behavior; as well as (c) thegeneral risk of the undertaking.

• The frequency of the transaction. TCE was originally devel-oped for repetitive, routine transactions, undertaken bythe classically managed organization in functional andhierarchical structures.

TCE proposes that high levels of asset specificity, uncer-tainty, and low frequency should lead to make decisions inorder to lower possible maladaptation costs. Low asset speci-ficity, uncertainty, and high frequency should lead to buy deci-sions in order to economize on existing scales and resources.In project environments TCE is especially appropriate duringthe presales phases of projects, up to contract award (Müller &Turner, 2005). The costs associated with transaction costs arelisted in the introduction section of this paper.


Transaction costs perspectives are typically applied atthe project level in order to economize on the costs of indi-vidual projects. In the case of higher frequencies of proj-ects—the case of multiple projects for an ERPimplementation within a buyer organization—the focusmay better be raised to the portfolio level, to economize ontransaction costs in the long term.

Control Perspective in ProjectsProject management is still widely understood as the processof planning and management in order to deliver in accor-dance with time, cost, and quality goals. The strict applica-tion of this traditional focus has been criticized extensively.Examples include Jugdev and Müller (2005), whose reviewof the evolving understanding of project success over the last40 years showed that these traditional success factors weresupplemented over time by a variety of other factors. Solefocus on the “golden triangle” can lead to rivalry amongprojects with an adverse affect on the results of all projects ina portfolio. Kerber and Buono (2003) argued that planningand control is a tool most suitable for bureaucratic organiza-tions. These organizations plan their work in well-definedplans of action and strictly follow-up and control plan adher-ence. They have a control perspective. Problems arise whenbureaucratic buyer organizations meet organic project teams.Here, the mutual expectations in terms of authority and com-munication are not fulfilled. Turner and Müller (2004b)showed empirically that project success is best achieved inmedium levels of structure, with neither too high, nor toolow levels of control. Consequently, an organic project team,working in a bureaucratic buyer organization could “e con-trario” give rise to conflicts between the project planning andthe need for knowledge creation.

However, planning is still necessary. One reason isshown by Mikkelsen and Riis (1989, p. 23), who highlightthat plans should give a color of control. Therefore, projectleaders should structure their project work through projectmethodologies and develop budget and time objectivesalready in the beginning of the project. However, the setobjectives are the basis for the contract, which is typicallysigned before the knowledge creation process and the proj-ect can be scoped in detail. A project plan at this presalesstage serves the purpose of selling the project and shouldconvince buyers as well as other stakeholders. Certainly, thebuying organization requires a ROI, NPV, or other measuresfor deciding on its investments. For example, this could bethe steering group, management group, or credit institutionsthat should admit loans or resources for the project. Thesteering committee in a company that has decided to imple-ment an ERP system and values a control perspectiveemphasizes bureaucratic processes for planning, organizing,leadership, and control (Alvesson, 2001).

The review showed that the shortcomings of focusingon time, cost, and quality as the main measures for projectsuccess are well known. However, these parameters still pre-vail for management thinking and decision-making oninvestments in projects.

Learning ProjectsOrlikowski (1993, 2000), Orlikowski and Barley (2001),and Orlikowski and Hofman (1997) argued that IT proj-ects could give rise to leveraging the learning during theproject implementation. However, Eskerød and Blichfeldt(2004) identified learning to be problematic, when proj-ect goal orientation, linear structure, and the time pres-sures do not allow for knowledge creation.

From a learning perspective Christensen and Kreiner(1997) described a learning project as one with knowl-edge development in focus. According to them, projectgoals can be seen as hypothetical and could be changeddepending on knowledge creation within the project. Thisis contrary to the evaluation of a project from a controlperspective, which is done by comparing current accom-plishments with the requirements defined at the begin-ning of a project. This control perspective can lead toconflicts (Christensen & Kreiner, 1997), and maladapta-tion costs (Müller & Turner, 2005). Similarly, Anderson(2006) argued that project deliverables should be definedand planned in small parts to be delivered over time andin accordance with the buying organization’s needs. Forinstance, they found that some projects lack time tosearch for help from other project leaders. In other words,these authors argued that project goals should be set earlyon in the project, but be revised as needed. Here, animplementation process is seen as an open learningprocess, taking into account the effects of the implemen-tation of the results on the buying organization. That willallow for learning and adaptation of deliverables to theneeds of the buying organization, allowing the learningprocess to be included.

The organizational structures could create otherbarriers for project learning, because they can affecttime, flexibility, and complexity (Sun & Scott, 2005).Propp, as cited by Sun and Scott, argued that projectteams do not take sufficient time for knowledge devel-opment due to time pressure. This constitutes a barrierfor knowledge development and can lead to suboptimalproject outcomes.

This is a result of focusing primarily on the transac-tion costs of the individual project to be accomplished,rather than on the sum of transaction costs for a series ofprojects over time and with the same buying organization.

The conflict of learning projects under traditionalproject management and from a TCE perspective was pre-viously identified. Based on the literature review, the fol-lowing propositions were developed:

Proposition 1: Knowledge development within learningprojects impacts transaction costs and can be economized inthe long-term at the portfolio level.

Proposition 1a: Suppliers’ learning perspectives in proj-ects with knowledge development impacts their transactioncosts in the short-term.

Proposition 1b: Suppliers’ learning perspectives in proj-ects with knowledge development impacts their transactioncosts in the long-term.


Methodology

A case study approach, following Yin (1994), was used forthe investigation of the connection between transactioncosts and knowledge creation within learning projects. Theproject followed was called project T and included theimplementation of a rather standardized ERP system.

Data collection was done through 10 open-ended inter-views and 11 participant observations in an ERP project overa period of six months from September 2005 to March2006. The project followed was defined by the supplierorganization as a learning project, where the transactioncosts were expected to be higher than regular ERP projects ofthe same type. Interviews took place with the CEO, projectmanagers, and line managers of the supplier organization,as well as line managers and a project leader of the buyingorganization. Observations were done in project meetingsbetween the buyer and supplier organizations with thesepeople participating. Interviews totaled 15 hours (approxi-mately 1.5 hours each), observations occurred over a periodof 180 days and 11 meetings. Interview data were tran-scribed and validated by the CEO of the supplier organiza-tion and two managers of the buyer organization.

Analysis of the data was done from Silverman’s per-spective (1993, p. 94) of interactionism, where “intervie-wees are to be viewed as subjects who actively constructthe features of their cognitive world. [...] In this ‘humanis-tic’ version of the interview, both the type of knowledgegained and the validity of the analysis is based on ‘deep’understanding.”

The results of the interviews and the observations wereinterpreted in the context and relationship of the sevenother projects in the portfolio with that customer. Thatinformation provided for portfolio-wide view of the impli-cations of an individual learning project at the project-port-folio level.

The results provided for both managerial as well as the-oretical implications, which are described in the conclu-sions section. The buyer and supplier organizations arepresented in the following sections.

Colour ABThe buyer company is a Swedish medium-sized wholesalerin special goods for the industry and small companies withabout 50 employees. Colour AB has the largest marketshare within this business in Sweden. They are also a retail-er for the industry and for consumers. The company is wellestablished and has been on the market for six decades. TheERP system to be implemented has been on the market forapproximately one year. It is installed at two sites of thecase company.

With this project, Colour AB decided to implement anERP system supporting the purchasing process. According tothe marketing manager, they also required an interface suit-able for Internet communication with their suppliers andcustomers. The size of the business did not admit invest-ments in a large server. Therefore, they wanted to join forceswith a supplier in a learning project and be the first compa-

ny with an ERP system programmed in Java language com-bined with a PC instead of a server. Their motivations wereto quickly improve effectiveness in their working processesthrough new technology, in order to survive in the market-place. Moreover their perspective was that Web-based inter-faces are future-oriented, because of the increasing numberof new employees who are computer savvy. Owing to thisthey decided to be first within their business to have thistype of solution.

Management wanted the system to be implementedwithin six months, based on a fixed-price contract withUDK AB as supplier. Consequently, if the project shouldrequire more resources than planned from the supplier,Colour AB should not pay for these resources.Management prepared their employees for the upcomingorganizational change.

UDK ABThe supplier has experiences with ERP projects since the1980s. Accordingly, the company is experienced in thesetype of projects. Their revenue in 2005 was approximately450,000 euro. Thus, the CEO could compare the project interms of transaction costs, control, and learning perspec-tives, to a number of already finished projects. UDK AB hasapproximately 25 employees working as consultants for theindustry and 10 of them are specialized in ERP systemimplementation. These systems are standardized and theprojects are implemented in cooperation between the cus-tomer and the supplier. UDK AB is responsible for the cus-tomization of the platform system so that it fulfills theagreed-upon requirements. UDK AB’s clients are responsiblefor the quality of the data within their existing databases.These data are transferred to the new system supplied byUDK AB.

The buyer and seller organizations’ resources have dif-ferent competencies within this area. To integrate the systeminto the organization, they have to communicate frequentlyand set clear objectives in order to satisfy mutual informa-tion needs.

The project was defined by the supplier as a learningproject, because of the new and unique combination of anew platform and hardware together with a JavaScript. In2005 this was a relatively new approach for ERP consul-tancies. Knowledge, competence, and experience neededfor these implementations were, according to the consult-ants, difficult to find in the market. According to the CEO,they had experience with all three of these technologies,but had not yet integrated them to the extent required forthis project. UDK AB encountered communication prob-lems, “translation barriers” (Andersson, 2006) among thesoftware, hardware, and the programming languages(Figure 1), as well as between the participants within theproject. Translation barriers are problems with understand-ing, for example, different languages. The participants camefrom different areas with different competencies within the“computer language” and the business as well. Further,integrating software, language, and hardware together


resulted in problems. The CEO had expected these techno-logical “barriers” and assigned more resources than usual tothe project, which were, eight consultants instead of theusual two.

In summary, UDK AB realized that the limited experi-ences from the implementations of these three separatetechnological parts needed knowledge development withinthe project and knowledge management within its organi-zation. This resulted in higher costs than planned for at thebidding stage. However, a learning perspective in project Tand partly in other projects, gives, according to the CEO,higher profits in further projects.

Analysis

This section describes and analyzes the supplier’s rational forthe handling of learning projects in the context of other proj-ects in the portfolio. Figure 2 illustrates the supplier’s differenttypes of ERP projects by their strategic perspective. Therefore,the upper level of the figure contains projects with a controlperspective, focusing on minimizing transaction costs. Themiddle level shows projects with new and strategically impor-tant customers. Here, the strategy is to nearly break even onprice versus costs in order to establish new business. Theseprojects have a combined learning and control perspective. Atthe lowest level are projects with a learning perspective wherehigher transaction costs are expected initially and thus profitsmay be compromised.

Figure 3 shows a scenario of a former strategic project atUDK AB, project F. This project had a break-even perspective.It was the first project with a new customer and thereforerequired knowledge development. In addition to project F,project U was developed as a consequence of learning aboutthe customer’s needs and organizational routines, similar towhat Andersson and Wilson (2006) described. Accordingly,UDK AB did not gain any profits from project F, but only fromproject U. After projects F and U, projects X, Y, and Z arose asnot yet contracted with the client. Figure 3 shows the classifi-cation of UDK AB’s different kinds of projects (Figure 3).

Figure 1: “Translation barriers” in technology

Figure 2: Project types and their possible occurrence over time

Figure 3: Strategic projects and their follow-on projects


In the case described in this paper, the supplier declaredproject T as a “learning project,” which led to additionalinvestments in the form of more resources from the suppli-er’s organization relative to other ERP projects, the size ofthe client organization, and the task to be solved.Accordingly, the learning perspective resulted in a less rigor-ous control of resources in terms of time and number ofresources in project T. This learning investment should beleveraged in future projects X, Y, and Z.

Analysis of the results from knowledge developmentwithin the contracted project T, shows that UDK AB’s classi-fication of project T as a learning project had implicationsfor transaction costs. Project T did not reach breakeven anda natural question was then why UDK AB ran this project atall. UDK AB’s portfolio decision could have been to focuson type L projects only, resulting in short-term profit maxi-mization due to lowest transaction costs (Figure 2).Contrarily, a strategic approach like that of project F wouldhave been possible and prevented losses in the first projectand created profits in project U, while having no impact onfuture projects (Figure 3).

In the case studied, project T, the supplier took a rela-tively large, but expected loss. The approach taken account-ed for losses in project T as an investment in learning forprojects X, Y, and Z. For all projects with Colour AB, the low-

est overall transaction costs and thus highest profits wereaccomplished by higher-than-usual upfront learning invest-ments as part of project T. Accordingly, UDK AB will apply acontrol perspective on future projects X, Y, Z in order tomaximize profits.

UDK AB earned higher profits in the long term as aresult of the knowledge created in the learning project(Figure 4). Knowledge development could therefore limitthe transaction costs in the long term. A strategic perspectivein case of project F led to only one profitable project in thelong term, while a learning perspective in project T led toprofits in several future projects, albeit at the expense of a30% plan overrun in the learning project.

Supplier’s Risk in Taking a Learning PerspectiveOne of the main risks arising for the supplier organization isthe buyer’s possible discontinuation of the project series. Thefinancial impact for the supplier is larger with a discontinua-tion at or immediately after the first project in the projectseries, provided there are no other projects where the learn-ing can be leveraged. Here, the supplier organization hasinvested substantially with the expectation of higher gains infuture projects. Figure 5 shows the risks associated with thatkind of situation, giving the three scenarios of control, strate-gic, and learning perspective from the short to the long term.

Figure 4: Impact of learning projects on classification of future projects

Figure 5: Supplier’s risk over time arising from buyer’s discontinuation


Taking a control perspective, the financial risk for thesupplier is lowest, as even the first project is designed to gaina profit. However, over the course of the project series, theoverall profit will be potentially lower among the three sce-narios because of the continuous investment in new learn-ing with every new project. The strategic perspective alsoprevents losses, but does not exploit the full potential oflearning at the beginning of the series of projects and there-fore also needs continuous, but potentially lower, invest-ments in learning with every new project. The learningperspective provides for the largest potential profits at thehighest risk. The return on investments in learning willprobably not be seen until the end of the second or eventhird project in the series. The supplier’s risk for financiallosses from a buyer’s discontinuation after the first projectmust therefore be addressed in the interaction between theparties. Mitigation strategies can be developed, among oth-ers, by the following:• Contractual clauses that allow for the (partial) recovery of

supplier’s investment in case of discontinuation within thefirst few projects

• Supplier providing increasingly favorable conditions/fees(or other contractual incentives) over the course of the firstfew projects

• Establishment of joint teams with high levels of collabo-ration, so that the investments are approximately equallyhigh for both parties, i.e., none of the parties benefits fromleaving the project.

In addition, the seller organization may leverage thelearning from the first project in similar projects with otherclients. However, the ultimate aim should be a long-termrelationship between the supplier and buyer with a steadyflow of projects over time.

Conclusion

This case study investigated the long-term transaction costsin a multi-project ERP system implementation at one buyerorganization. A case study with 10 interviews and 11 partic-ipant observations over six months was used to develop amodel outlining the differences between short and long-term transaction costs at portfolio level, accruing from theneed to learn about the buyer firm’s organization. Thesewere as follows:• Short-term, non-strategic approaches—profit maximization• Strategic approaches to learning in projects without finan-

cial losses—breakeven expenditures• Learning approaches—short-term loss with long-term profit.

Proposition 1 was supported by the case. Defining a proj-ect as learning project at the outset, with a long-term view ofunderstanding the client’s organization in detail to solve prob-lems within future projects, paid off for the case company.

Proposition 1a was supported. The impact is in form oflosses in the first project, which will be recovered in thelong-term. In the case of UDK AB, the losses amounted to30% plan overrun.

Proposition 1b was supported. The impact was positivefor UDK AB. The investment in the learning project paid offin the long-term through lowest possible transaction costsin the projects following the learning project.

The research question can now be answered in terms ofmanagerial and theoretical implications.

Managerial ImplicationsIn reference to the strategic and learning perspective previ-ously outlined, the following is suggested through the study:1. Suppliers should scope the chances and number of possi-

ble future projects with the client organization.2. A “higher-level” portfolio perspective should be taken

when estimating the learning needs for future projects. Thelearning investments should be scoped for the foreseeablebusiness opportunities with the client organization.

3. Resources should be provided for learning, including anevaluation of the possible opportunity costs for not hav-ing the resources at other, more profitable, projects.

4. In cases where a supplier expects relatively few ERP proj-ects with one client to occur over time, the supplier organ-ization should decide for a strategic perspective andaccept a near to break-even result in the first project, fol-lowed by profits in the following project. This is achievedthrough the containment of the learning efforts to theshort-term needs, not taking into account the learningneeds for future projects. Transaction costs are not mini-mized in the long term. However, risks of losses (e.g.,through the client abandoning the business after the firstproject) are minimized.

5. In cases where a supplier expects a larger number of ERPprojects with one client over time, the supplier organiza-tion should decide for a learning approach and invest inlearning about the client’s organization, over and above theneeds of the first project, to leverage this knowledge in thelong term. Thus, run all future projects profitably.Transaction costs are minimized in the long term.However, risks of losses through the client abandoning thebusiness are high in the beginning, but diminish over time.

Theoretical ImplicationsThe TCE perspective showed a need to manage learning costsat the project-portfolio level, with a long-term view in mind.It supports Williamson’s (1985) argument that transactioncosts in reoccurring transactions are best done in the market,leverage existing scales and resources, and give an advantageto the first supplier in the series of the transactions. This isbalanced by the level of uniqueness of the individual projects(i.e., their asset specificity), which should be moderate incase of a series of ERP projects in one buyer organization.

The learning approach minimizes the supplier’s mal-adaptation costs by building a profound and comprehensiveunderstanding of the buyer organization at the early stagesof the project series.

The strength of the study lies in its use of a well-experi-

enced and long-established case company and the information


from the strategic leader—the CEO. Weaknesses are in the

small sample, that is, one case only. Further research is need-

ed to validate the results of the study and refine the findings

in terms of business situations, geographies, etc. To that end,

a quantitative study, with a larger number of ERP suppliers

should be used and the results triangulated.TCE has again shown its applicability in project-based

environments. This time the view was raised from projectlevel to portfolio level. Applying TCE thinking at the portfo-lio level should be beneficial in the long-term for project-oriented companies.

References

Alvesson, M. (2001). Organisationskultur och ledning(Sweden, Malmö, Liber).

Andersen, E. (2006). Perspectives on projects.Proceedings of the PMI Research Conference, Montreal, Canada.Newtown Square, PA: Project Management Institute.

Andersson, A., & Wilson, T. (2006, June). CustomizedERP projects: Toward understanding their sequential nature.Projects and Profits.

Bajaj, J. (1997). Analyses of contractors, approaches torisk identification in New South Wales Australia.Construction Management and Economics, 15, 363–369.

Christensen, S. & Kreiner, K. (1997). Projektledning: Att ledaoch lära i en ofullkomlig värld. Academia Adacta, Lund, Sweden.

Eskerød, P., & Blichfeldt, B. S. (2004). Do implicit coreassumptions of project management community culturerestrain intra-organizational learning across project man-agers? In Wikström and Artto (Eds.), Proceedings for theIRNOP VI Conference (Finland: Åbo, Åbo AkademiUniversity and Helsinki University of Technology).

Jiang, J. J., Klein, G., Chen, H.-G. & Lin, L. (2002).Reducing user-related risks during and prior to system devel-opment, International Journal of Project Management, 20(7),507–515.

Judgev, K., & Müller, R. (2005). A Retrospective look atour evolving understanding of project success. ProjectManagement Journal, 36(4), 19–31.

Kerber, K., & Buono, A. F. (2003). A Situational Modelof Organizational Change: Complexity, Uncertainty andApproaches to Change. In Koppel, Nicole (Ed.), Proceedingsof the Northeast Business & Economics Association, pp.281–283. Montclair, NJ: Montclair State University.

Kumar, R. L. (2002). Managing risks in IT-projects: Anoptions perspective. Information and Management, 40,63–74.

Kwak, Y. H., & LaPlace, K. (2005). Examining risk toler-ance in project-driven organizations. Technovation: AnInternational Journal of Technical Innovation andEntrepreneurship, 25(6), 691–695.

Lind, N. (2002). Time effects in criteria for acceptable

risk. Reliability Engineering and System Safety, 78, 27–31.Linderoth, H., & Lundqvist, A. (2004). Learn not to

learn: Paradoxical knowledge creation and learning in ERP-projects. Proceedings IRNOP VI, (pp. 409–422). Turku,Finland: Academy University Press.

Mikkelsen, H., & Riis, J. O. (1989). Grundbog i projek-tledelse. Holte, Denmark.

Mills, A. (2001). A systematic approach to risk manage-ment for construction. Structural Survey, 19(5), 245–252.

Müller, R., & Turner, J. R. (2005). The impact of princi-pal-agent relationship and contract type on communicationbetween project owner and manager. International Journal ofProject Management, 23(5), 398–403.

Orlikowski, W. J., & Barley, S. R. (2001). Technology andinstitutions: What can research on information Technologyand research on organizations learn from each other? MISQuarterly,. 25(2), 145–166.

Orlikowski, W. J., & Hofman, J. D. (1997). An improvi-sational model for change management: The case of group-ware technologies. Sloan Management Review, Winter.

Orlikowski, W. J. (1993, Sept.). Case tools as organiza-tional change: Investigating incremental and radical changesin systems development. MIS Quarterly, 17(3).

Orlikowski, W. J. (2000). Using technology and consti-tuting structures: A practice lens for studying technology inorganizations. Organization Science, 11(4), 404–428.

Patterson, F. D., & Neailey, K. (2002). A risk registerdatabase system to aid the management of project risk.International Journal of Project Management 20(20), 365–374.

PMI. (2004). A guide to the project management body ofknowledge (PMBOK® guide). Newtown Square, PA: ProjectManagement Institute.

PMI. (2006). The standard for portfolio management.Newtown Square, PA: Project Management Institute.

Silverman, D. (1993). Interpreting qualitative data.London, UK: Sage Publications.

Sun, P. Y-T., & Scott, J. L. (2005). An investigation ofbarriers to knowledge transfer. Journal of KnowledgeManagement, 9(2), 75–90.

Turner, J. R., & Keegan, A. (1999). The versatile project-based organization: Governance and operational control.European Management Journal, 17(3), 296–309.

Turner, J. R., & Müller, R. (2003). On the nature of theproject as a temporary organization. International Journal ofProject Management, 21(1), 1–7.

Turner, J. R., & Müller, R. (2004a). A comparison oftransaction costs and agency costs on projects. Proceedings ofthe 18th World Congress on Project Management. InternationalProject Management Association, Budapest, Hungary.

Turner, J. R., & Müller, R. (2004b). Communication and co-operation on projects between the project owner as principal andthe project manager as agent. European Management Journal, 22(3).


Turner, J. R. (1999). The handbook of project-based man-agement—Improving the processes for achieving strategic objec-tives. Berkshire: McGraw-Hill.

Williamson, O. E. (1975). Markets and hierarchies:Analysis and antitrust implications. New York: CollierMacmillan, Canada, Ltd.

Williamson, O. E. (1985). The economic institutions ofcapitalism. New York: The Free Press.

Williamson, O. E. (1996). The economic institutions ofcapitalism. New York: The Free Press.

Yin, R. K. (1994). Case study research: Design and meth-ods, 2nd ed. London: SAGE Publications Ltd.

ANNIKA ANDERSSON is a PhD student and lecturer at Umeå School of Business, Umeå University, Sweden. She lectures

in project management, human resource management, business-to-business, and relationship marketing. Her research

is on relationship issues in IT-consultancy firms. Prior to her academic career she spent 25 years in business, mainly

working for an insurance company. Her projects were used to implement IT support for the business processes and

activities, and knowledge creation was enforced on a daily basis. Annika Andersson is the editor of an upcoming case

book for teachers, using ERP systems as a pedagogical tool. She is also a board member at SANTE – Scandinavian

Association Network for Teaching Enterprise systems.

RALF MÜLLER, DBA, is associate professor at Umeå University in Sweden and visiting professor at ESC Lille in France. He

lectures in project management, governance of project-based organizations, and research design and methodology. His

current research focus is on program and portfolio management, and the interactive effects of project managers’

leadership style with project type on project success. His past research was published in European Management Journal,International Journal of Project Management, and the Project Management Journal. The majority of his 30 years in business

were spent consulting for large enterprises in project management and governance (e.g., as worldwide director of project

management in NCR Teradata). Dr. Müller is a reviewer for the International Journal of Project Management and the ProjectManagement Journal, author of three books and a number of articles in international project management journals, as

well as a frequent speaker at researcher and practitioner conferences. He is a co-founder of two PMI chapters in Europe

and contributor to the following PMI standards: Organizational Project Management Maturity Model (OPM3®), Standardfor Program Management, and Standard for Portfolio Management.


PROJECT MANAGEMENT RESEARCH—THE CHALLENGE AND OPPORTUNITY

Project management is one of the fastest

growing disciplines in organizations

today. However, ironically, the statistics

of project success suggests that most

projects still fail and many projects do not

accomplish their business results. This

presents possibly a unique opportunity

for substantial improvement. In this

paper the authors offer their perspective

about the challenge that the project man-

agement research community is facing

today. The authors propose several

research directions that may evolve as

central in the next few years in order to

stimulate the discussion and debate

about the future of the discipline. They

first look at project management research

from a problem-driven perspective and

than offer three central views with which

project management could be perceived:

the strategic/business view, the opera-tional/process view, and the team/leader-ship view. For each one the governing

thought pattern, the theories, and the

related disciplines are presented.

Although these views are certainly not

unique, they may provide an integrated

perspective of the discipline and a possi-

ble trigger for further discussion that may

help attract scholars from other more

established academic disciplines and

improve the status of project manage-

ment research.

Keywords: project management; research;strategic business; operational; leadership


Introduction

In a rapidly changing world, the magnitude of project activity in organizationsis constantly growing. Projects are used today, not just for new product devel-opment or construction, but also for numerous other reasons, such as product

improvement, system deployment, process building, process reengineering, newservice initiation, software development, marketing campaigns, and many others.Furthermore, today’s dynamic environment makes product lifecycles much short-er, forcing organizations to accelerate new product development and start manymore new projects. Even stable industries such as banking, retail, or insurancethat were used to stick to their traditional business processes, find that they needmore and more new initiatives, and that those can only be achieved through proj-ects. Finally, the information technology (IT) and Internet revolution is far fromslowing down, forcing almost every organization to invest in new IT projects tokeep up with demand and competition.

Yet, in a paradoxical way, project failures, delays, and disappointments aremuch too common to be ignored. With most projects failing to meet time andbudget goals, and many projects are not fulfilling their business objectives, thereseems to be an alarming gap between the needs of the discipline and what weknow in order to fix them. From a research perspective, there is a great opportu-nity to help close this gap.

In this paper, some observations on the challenges and opportunities in proj-ect management research today are provided. The goal is to stimulate the discussionabout the future of project management research and about the role that thisresearch could play in academia, and industry. The authors must note, however, thattheir observations are more subjective than universal, and in no way are theyexhaustive nor could they preclude different views or other avenues for exploration.

The Evolution of Project Management

Since the beginning of civilization, people have been involved in organized cre-ations that have influenced lives and societies. Today, we would call them proj-ects. Such creations often left impressive monuments and legendary stories thathave inspired future generations and motivated the research of historians, sociol-ogists, archeologists, and management scholars. From the Roman Aqueducts tothe Great Wall of China to the Egyptian Pyramids, humanity has always been fas-cinated by the ingenuity and creativity of the human mind and by the power and

ABSTRACT

AARON J. SHENHAR, Stevens Institute of TechnologyDOV DVIR, Ben Gurion University, Israel


analyze future business, technology,and environment trends.

A different example was the SydneyOpera House construction project:

One of the world’s greatest touristattractions is the Sydney OperaHouse, an architectural wonder visit-ed every year by millions of travel-ers.3 The original project plan asenvisioned by the New South Walesgovernment in the 1950s includedan estimated budget of about $7million (Australian) and a scheduleof five years. But getting there wastough. The construction projectexperienced enormous difficulties—extensive delays, bitter conflicts, andpainful budget overruns. More than16 years passed before the OperaHouse opened its doors, and its finalprice tag was over $100 million!

Assessing the project based onefficiency measures only; namely,on its time and budget performance,you might conclude that the SydneyOpera House was a classic exampleof project management failure.Today, however, no one really careshow the project was managed, andalmost everyone sees the OperaHouse as a success story. It providescontinuous income and fame to theCity of Sydney and it remains one ofthe most fascinating buildings inthe world.

A third relevant example is the construc-tion of Denver International Airport:

Denver International Airport wasinitiated as a four-year project in1989 in order to take over Denver’sgrowing transportation needs fromStapleton Airport, which had out-grown its maximum capacity.4 As aconstruction project, nobody reallyexpected any particular problems. Itwas planned and executed as anyother large project, using all avail-able tools for planning and control.Yet the program suffered an exten-sive delay of 16 months and a costoverrun of $1.5 billion. As it turnedout, one component of this project,

ability to organize, coordinate, andexecute collective people’s efforts.

The authors define a “project” as atemporary organization and process set upto achieve a specified goal under the con-straints of time, budget, and otherresources, and “project management” asthe managerial activities needed to lead aproject to a successful end.1 As a formalmanagerial discipline, project manage-ment is still relatively young. It has for-mally evolved in the middle of the20th century, when the first programevaluation and review technique(PERT) marked the beginning of a newdiscipline.2 With time, tools, tech-niques, and methods became standar-dardized across industries, andbusinesses, and more and more organ-izations began witnessing the benefitsof organizing work around projects.The formation of the ProjectManagement Institute (PMI) in 1969as the leading professional associationwas an important landmark in thedevelopment of the discipline. A Guideto the Project Management Body ofKnowledge (PMBOK® Guide) – ThirdEdition became the premier standardon how to manage today’s projects(PMI, 2004).

What is Missing?

In spite of civilization’s long history ofmonumental projects, and half a cen-tury of building the discipline, projectperformance is alarmingly low. Studyafter study has shown that most proj-ects are not completed on time, areover budget, and even when complet-ed, many do not meet management orcustomer expectations. For example,according to a Standish Group report,only 28% of projects are successful,23% are failures, and 49 % are provid-ing only partial answers, with an aver-age fulfillment of 67% of the featuresdefined at project initiation (StandishGroup, 1994, 2001). In other studies, itwas shown that 85% of the projectscome short of time and budget goals,with an average overrun of 70% intime and 60% in budget (Shenhar &

Dvir, 1993). Almost all other studieshave confirmed the same or evenworse results.

The current project managementbody of knowledge is rich and helpful. Itforms the foundation for basic trainingand learning project management andclearly, anyone who wants to acquire theknow-how in this profession must learnthe basics. Thus, many of the traditionalproject management techniques becamestandard and were transformed intonumerous tools and applications, withincreased detail and sophistication. Butas project results demonstrate, all thismay not be enough.

Clearly, some project failure maystem from managerial neglect or lack ofplanning. But as the evidence suggests,sometimes even well-managed projectsfail to meet their objectives, and some-times, even poor planning may notnecessarily lead to a project’s failure.Consider the following few cases:

The Iridium Project wasMotorola’s $5 billion project aimed toprovide global communication servicevirtually anywhere at anytime(Barboza, 2000; Carayannis, Anbari, &Kwak, 2003). The project was carefullyplanned and managed but it turnedout to be a business nightmare.

In November 1998, the Iridium net-work was established and started toprovide global network services.However, in March 2000, Iridiumfiled for bankruptcy terminating itsservices. The project was onceviewed as a technological break-through; yet, it ended quickly andmysteriously. The program officewas established with full-time proj-ect control managers, and relocatedsoftware engineers and analysts. Inaddition, the control managers uti-lized sophisticated project manage-ment software, Primavera ProjectPlanner, to handle all the complexand interrelated project schedulingproblems (Fabris, 1996). In retro-spect, however, although theIridium project was well-managedaccording to the book, it failed to

1 Notice that the authors’ definition is different from the Project Management Institute’s, which defines a project as “a temporary endeavor undertaken to create a unique product or service,”and project management as “the application of knowledge, skills, tools, and techniques to project activities in order to meet project requirements” (PMI, 2004).

2 Peter Morris provided an excellent overview of the history of project management (1994). See also Carayannis et al. (2003).

3 A short review of the Sydney Opera House construction project can be found at: http://www.aviewoncities.com/sydney/operahouse.htm; see also Murray (2003).

4 The full story of the Denver Airport can be found in the Harvard Business School Case, November 6, 1996. BAE Automated Systems, Implementing the Denver International Airport Baggage-Handling System (A) and in Dempsey, Goetz & Szyliowicz (1997).


the automatic bag-handling system,had a higher risk than the rest, but itwas treated as a standard, well-proven subsystem, just like anyother part of the project. The baghandling system was the main rea-son for the huge budget overrun anddelay in opening the airport.

These projects are not unique.Similar examples may be found inalmost every organization (Shenhar &Dvir, 2007). Perhaps there is more toproject success than following a speci-fied set of rules. Perhaps it is not thetools or applications, nor is it the lackof process. Project management todayhas plenty of those, and while impor-tant, they may be not enough.5 Theconclusion here is that the problem ismuch deeper. It is at the conceptuallevel, rather than in process or prac-tice.6 The authors believe it is time todevelop new understanding of whatproject management is all about. Suchunderstanding will most likely have animpact on the practice and educationin the discipline, and eventually influ-ence the development of tools andprocesses. This is the responsibilityand challenge of research.

The Research Challenge

Kloppenborg and Opfer (2002) provid-ed a detailed review of project manage-ment research, covering more than 40years of publications. According totheir observations, project manage-ment research was focused on planningand scheduling during most of the1960s. In the 1970s, automated soft-ware for project management has creat-ed an increased interest in cost andschedule control. This trend continuedinto the 1980s, with new studies onlife-cycle costing and risk managementplanning. Yet that time also marked theappearance of studies on team buildingand leadership, leading to the 1990s,with even more focus on humanresources, teams, and leadership.

A new stream of studies emerged,however, in the 2000s. Those included

research on project typologies andcontingencies (Crawford, Hobbs &Turner, 2004; Pich, Loch & De Meyer,2002; Shenhar, 2001; Shenhar & Dvir,2004; Turner & Muller, 2003), strategicproject management (Cleland &Ireland, 2002; Judev, 2003; Morris,2004; Shenhar, 2004), and the global-ization of projects (Armstrong, 2000;Artto & Dietrich 2004; Cairncross,2002; Kruglianskas & Thanmhain,2000). Yet, no central paradigm hasemerged so far that is underlying theresearch and conceptualization ofproject management or is influencingthe practice of project management.

Perhaps an answer could befound in other fields. Conceptually,there may be other closely relatedfields that have evolved at the sametime as project management, andwhere theories and research had ahigher impact. A brief look at some ofthese disciplines may suggest the fol-lowing observation.

Technology and InnovationManagement ResearchTechnology and innovation research isprobably a more influential and theo-retically developed discipline, coveringa wide range of literature. It startedwith the early studies of Burns andStalker (1961) that introduced a classi-cal distinction between incrementaland radical innovation. And it contin-ues today with varying points of viewand lessons that were applied to realorganizations. Typical studies dis-cussed the process of innovation, struc-tural, architectural and systeminnovation, innovative cultures, andthe theory of disruptive technologies(e.g., Anderson & Tushman, 1990;Brown & Eisenhardt, 1997;Christensen, 1997; Christensen &Raynor, 2003a; Eisenhardt & Tabrizi,1995; Gatignon, Tushman, Smith, &Anderson, 2002).

New Product Development ResearchA second rich direction is the newproduct development. Here researchers

explored among other things, the criti-cal success factors, the developmentstage process, and the influence of theenvironment on these processes(Brown & Eisenhardt, 1995; Cooper,1994; MacCormack, Verganti, &Iansiti, 2001).

The Entrepreneurship LiteratureYet another relevant direction is thestudy of entrepreneurship. Here studieshave focused on the process of idea gen-eration, crafting new businesses, andtranslating ideas into practice in organi-zations (e.g., Shane & Ulrich, 2004).

Operations ManagementFinally, there is the well-establishedfield of operations management. Herestudies looked at projects as processesand tried to apply optimization tech-niques to improve project perform-ance. Critical chain is probably themost well-known approach that wasadopted to project management(Goldratt, 1997).

Ironically, however, althoughclose in nature and often addressingsimilar problems, very few of thesedirections have had a significantimpact on the discipline and practiceof project management. These observa-tions, we believe, suggest that today theresearch community in project man-agement is facing a rare opportunity toexploit a wide spectrum of new direc-tions, which may address the currentneeds of the discipline.

Opportunity, however, is associat-ed with risk; therefore, listed first are afew possible risks:• There are no commonly accepted par-

adigms that are leading the perceptionand understanding of project man-agement. There are no central theoriesthat are underlying the discipline.

• Although project management isconsidered to be an interdisciplinaryfield, only a few studies haveattempted to present the interdisci-plinary approach and its applicationto project management. And onlyfew researchers from other fields

5 Morris (2002) claimed that what many practitioners are now looking for, particularly those charged with developing project management competencies within companies, is at least some evidenceto show that the application of formal project management knowledge and practices produces better project outcomes. Yet the current data on this is only slight (Crawford, 2001; Ibbs & Kwak,1997; Morris, 1987, 2001; Pinto & Slevin, 1988, 1989). There is no study yet that demonstrates a causal relationship between the application of formal project management and project outcomes.

6 For example, project success factors are contingent upon the specific type of project and the list of project success factors incorporated in the large bodies of knowledge is far from universal.Project management has a wide range of variations and projects have fewer characteristics in common than previously considered. Therefore, adhering to the common set of management procedures and techniques can not guarantee projects success (Dvir, Lipovetsky, Shenhar, & Tishler, 1998).


were drawn to study project manage-ment related issues.

• Project management is still not recog-nized as an equally important mem-ber among the more establishedresearch disciplines of management,such as marketing, operations man-agement, finance, or strategy.Similarly, there are very few groups ordepartments of project managementin business or management schools.Finally, project management researchis rarely found in the leading aca-demic management journals, whilethe journals dedicated purely to proj-ect management research are not rec-ognized among the highly prestigiousacademic outlets.

• Much of the project managementtraining is built on hands-on experi-ence and tool-practicing. There is lit-tle evidence that the researchliterature has had a substantialimpact on project management edu-cation, even at the academic educa-tional level.

Given the challenges just men-tioned, it may be the right time todevelop a wider research agenda thatwill address these challenges and bringproject management research to theforefront of the academic world.However, to conduct research, scholarshave always looked for theory withwhich they can view the world and testtheir hypotheses. Furthermore, theoryis not only needed to study andexplain the real world, it may also helppractitioners and educators betterunderstand situations and conse-

quences. By using theory, one can gen-eralize the lessons from case studiesand better predict what may work ornot, and in what case (Christensen &Raynor, 2003b).

In the next sections two perspec-tives for future research is suggested—the problem-driven perspective andthe central paradigm perspective.

The Problem-Driven Perspective

Project management is a problem-driv-en discipline. It often deals withimportant questions and improvementin society, economy, and business.Currently, much of the learning inproject management is based on les-sons learned, experience, and bestpractices, and a great portion of projectmanagement education is based ontools and applications. Furthermore,even case studies are relatively missing.For example, only 2% of the 7,000Harvard Business School case studycollection mention projects and only afew dozen are actually dealing withproject management issues.

Yet, as mentioned, current projectmanagement concepts and theoriesare scant, and the discipline suffersfrom a lack of concepts and frame-works. However, theory may evolvefrom practical questions and real-lifeproblems. Such problems may promptscholars to search for a theory to dealwith the problem. Table 1 offers ashort sample of typical project man-agement practical problems and thetheories that could emerge to addressthese problems. No doubt there aremany more options.

The Central Paradigm Perspective

Is there a central paradigm in projectmanagement? By a central paradigmwe mean the lenses through which onemay view the field. Although otheroptions are possible, we suggest seeingresearch according to three major viewsof what project management is all about.The first and most traditional view seesa project as a sequence of activities thathave to be performed and completedaccording to plan, referred to here asthe operational/process view. The secondview is based on team/leadership. Itlooks at projects as an organizationalteam that needs to be led and motivat-ed. The third view, the strategic/businessview sees projects as business-relatedactivities that need to achieve the pro-ject’s business results.

Each view is completely different.Each is based on different assumptions,uses different metrics of success, anddefines the project manager’s role in adifferent way (see Table 2). The threeviews are described in Table 3 withtheir major research implications—thetheories, the possible methods, andperhaps most important—what are theother disciplines that can contribute toadditional views to studies in projectmanagement.

The Challenge of Integrative Research—

Attracting Researchers From Other Fields

Although each direction is a world of itsown, the real challenge is to combinethem all into a unified view. Success inproject management can only beachieved by an integrated holistic viewon the entire landscape of the project.

Table 1: Problem-driven theory development


From a research perspective, this meansmultidisciplinary research. It representsa unique challenge mostly to moreexperienced researchers, who estab-lished their name in a single discipline,but are now ready to exploit an inte-grated point of view.

Similarly, given the high potentialand impact of projects, the authorsbelieve other researchers may be ready tocontribute to the discipline of projectmanagement. There are plenty of fieldsthat can provide experience, methods,and new views to project management.To mention a few: marketing, operations,MIS, economy, human resources, leader-ship, quality, systems engineering, etc.

The Outlet Challenge—Publishing in

Leading Journals

Universal academic standards encour-age researchers to publish in leading

research journals. Typically the criteriafor the top journals are the rate ofacceptance and citation, as well asscholars’ perceptions. Kloppenborgand Opfer’s (2002) survey has shownthat less than 3% of the 3,000 projectmanagement studies that were pub-lished in more than four decadesappeared in the leading managementjournals. To establish project manage-ment among the traditional manageri-al academic disciplines, more projectmanagement studies are needed in therecognized leading research journals.

Table 4 includes the list of themost prestigious journals in man-agement and a list of leading jour-nals in technology, innovation,and engineering management thatare a natural home for projectmanagement research (Linton &Thongpapanl, 2004). Perhaps the

first step could be to position theproject management journalsamong the list of the leading tech-nology and innovation manage-ment journals.

Conclusion

Project management research is stillevolving. It has not yet established itsrole among the more traditional aca-demic disciplines of managementsuch as, marketing, finance, and oper-ations. Perhaps, as claimed, the reasonis the lack of a strong theoretical basisand a guiding set of concepts. Theauthors hope that the ideas presentedhere will prompt additional theoreti-cal development and further discus-sion. Clearly, project management isfar too complex to be explained byone unified theory. Yet, additional the-ories and concepts will attract more

Table 2: Major views of project management

Table 3: Theoretical and research implications


scholars and encourage more research,both of which will help establish thediscipline within academia.

Project management research hasgreat potential. Much of today’s orga-nizational activity is wrapped up inprojects and it only keeps growing.The field is rich with challenges. Withmore young researchers pursuingcareers in project managementresearch, and more resources beingallocated to research, the future ofproject management as a scholarlypursuit appears promising.

References

Anderson, P., & Tushman, M.(1990). Technical discontinuities anddominant designs: A cyclical model oftechnological change. AdministrativeScience Quarterly, 35, 604–633.

Armstrong, D. (2000). Buildingteams across boarders. ExecutiveExcellence, 17(3), 10.

Artto, K. A., & Dietrich, P. H.(2004). Strategic business manage-ment through multiple pjects. In P. W.G. Morris & J. K. Pinto (Eds.), TheWiley Guide to Managing Projects (pp.144–176). Hoboken, NJ: John Wiley &Sons, Inc.

Barboza, D. (2000, April 11).Iridium, bankrupt, is planning a fiery end-ing for its 88 satellites. New York Times.

Bass, B. M. (1999). Two decades ofresearch in transformational leader-ship. European Journal of Work and

Organizational Psychology, 10, 9–32.Brown, S. L., & Eisenhardt, K. M.

(1995). Product development: Pastresearch, present findings, and futuredirections. Academy of ManagementReview, 20, 343–378.

Brown, S. L., & Eisenhardt, K. M.(1997). The art of continuous change:Linking complexity theory and time-paced evolution in relentlessly shiftingorganizations. Administrative ScienceQuarterly, 42, 1–34.

Burns, T., & Stalker, G. (1961). TheManagement of Innovation. London:Tavistock.

Cairncross, F. (2002). The companyof the future. Cambridge, MA: HarvardBusiness School Press.

Carayannis, E. G., Anbari F. T., &Kwak, Y-H. (Eds.). (2003). The story ofmanaging projects: An interdisciplinaryapproach. Quorum Books.

Christensen, C. (1997). The inno-vator’s dilemma. Boston: HarvardBusiness School Press.

Christensen, C., & Raynor, M. E.(2003a). The innovator’s solution—Creating and sustaining successful growth.Boston: Harvard Business School Press.

Christensen, C., & Raynor, M. E.(2003b). Why hard nosed managersshould care about management theo-ry? Harvard Business Review, September,67–74.

Cleland, D. I., & Ireland, L.(2002). Project management: Strategicdesign and implementation (4th ed. vol.

1). New York: McGraw-Hill.Cooper, R. G. (1994). Third-genera-

tion new product processes. Journal ofProduct Innovation Management, 11, 3–14.

Crawford, L. (2001). Profiling thecompetent project manager in projectmanagement research at the turn of themillennium. In D. P. Slevin, D. I.Cleland, & J. K. Pinto (Eds.), Proceedingsof PMI Research Conference 2000.Newtown Square, PA: ProjectManagement Institute.

Crawford, L., Hobbs, J. B., & Turner,J. R. (2004). Project CategorizationSystems. Newtown Square, PA: ProjectManagement Institute.

Dempsey, P. S., Goetz, A. R., &Szyliowicz, J. S. (1997). DenverInternational Airport: Lessons learned.New York: McGraw Hill.

Dvir, D., Lipovetsky, S., Shenhar,A., & Tishler, A. (1998). In search ofproject classification: A non-universalapproach to project success factors.Research Policy, 27, 915–935.

Eisenhardt, K. M., & Tabrizi, B. N.(1995). Accelerating adaptive process-es: Product innovation in the globalcomputer industry. AdministrativeScience Quarterly, 40, 84–110.

Fabris, P. (1996, April 1). Groundcontrol. CIO Magazine.

Gatignon, H., Tushman, M.,Smith, W., & Anderson, P. (2002). Astructural approach to assessing inno-vation: Construct development ofinnovation locus, type, and characteris-tics. Management Science, 48(9),1103–1122.

Goldratt, E. M. (1997). The criticalchain. Avebury.

Ibbs, C. W., & Kwak, Y-H. (1997).The benefits of project management:Financial and organizational rewards tocorporations. Newtown Square, PA:Project Management Institute.

Jugdev, K. (2003). Developing andsustaining project management as a strate-gic asset: A multiple case study using theresource-based view. Unpublished PhDdissertation. University of Calgary.

Kloppenborg, T. J., & Opfer, W. A.(2002). The current state of projectmanagement research: Trends, inter-pretations, and predictions. ProjectManagement Journal, 33(2), 5–18.

Kruglianskas, I., & Thamhain, H.

Table 4: The most prestigious journals in management and leading journals in technology and

engineering management


(2000). Managing technology-basedprojects in multinational environ-ments. IEEE Transactions on EngineeringManagement, 47(1), 55–64.

Linton J. D., & Thongpapanl, N.(2004). Perspective: Ranking the tech-nology innovation management jour-nals. Journal of Product InnovationManagement, 21, 123–139.

MacCormack, A., Verganti, R., &Iansiti, M. (2001). Developing prod-ucts on “Internet time:” The anatomyof a flexible development process.Management Science, 47(1), 133–150.

Morris, P. W. G., & Hough, G. H.(1987). The anatomy of major projects.New York: John Wiley.

Morris, P. W. G. (1994). The man-agement of projects. London: ThomasTelford.

Morris, P. W. G. (2001). Updatingthe project management bodies ofknowledge. Project Management Journal,32(3), 21–30.

Morris, P. W. G. (2002). Science,objective knowledge, and the theory ofproject management. Proceedings of theInstitution of Civil Engineers, CivilEngineering, 150(2), 82–89.

Morris, P. W. G. (2004). Movingfrom corporate strategy to project strat-egy: Leadership in project manage-

ment. In D. P. Slevin, D. I. Cleland, &J. K. Pinto (Eds.), Innovations: Projectmanagement research. Newtown Square,PA: Project Management Institute.

Murray, P. (2003). The saga ofSydney Opera House: The dramatic storyof the design and construction of the iconof modern Australia. Spon Press.

Pich, M. T., Loch, C. H., & DeMeyer, A. (2002). On uncertainty,ambiguity and complexity in projectmanagement. Management Science,48(8), 1008–1023.

Pinto, J. K., & Slevin, D. P. (1988).Critical success factors across the proj-ect life cycle. Project ManagementJournal, 19(3), 67–75.

Pinto, J. K. & Slevin, D. P. (1989).Project success: definitions and meas-urement techniques. Project ManagementJournal, 19(1), 67–75.

PMI. (2004). A guide to the projectmanagement body of knowledge (3rd edi-tion). Newtown Square, PA: ProjectManagement Institute.

Shane, S., & Ulrich, K. (2004).Technological innovation, productdevelopment, and entrepreneurship inmanagement science. ManagementScience, 50(2), 133–144.

Shenhar, A. J. (2001). One sizedoes not fit all projects: Exploring clas-

sical contingency domains.Management Science, 47(3), 394–414.

Shenhar, A. J. (2004). Strategicproject leadership: Toward a strategicapproach to project management.R&D Management, 34(5), 569–578.

Shenhar, A. J., & Dvir, D. (1993).Managing R&D defense projects. Tel-AvivUniversity, Institute for BusinessResearch and Ministry of Defense, Israel.

Shenhar, A. J., & Dvir, D. (1996).Toward a typological theory of projectmanagement. Research Policy, 25,607–632.

Shenhar, A. J., & Dvir, D. (2004).How projects differ and what to doabout it. In J. Pinto & P. Morris (Eds.),Handbook of managing projects. NewYork: Wiley.

Shenhar, A. J., & Dvir, D. (2007).Reinventing project management: The dia-mond approach to successful growth andinnovation. Boston: Harvard BusinessSchool Press.

Standish Group. (1994). The chaosreport. The Standish Group.

Standish Group. (2001). The chaosreport. The Standish Group.

Turner, J. R., & Muller, R. (2003).On the nature of the project as a tempo-rary organization. International Journal ofProject Management, 21(1), 1–8.

AARON J. SHENHAR is the Institute Professor of Management and the founder of the project management program at

Stevens Institute of Technology, the Howe School of Technology Management. He has also been a visiting professor at the

University of Minnesota and Tel-Aviv University. He is the first winner of the Project Management Institute Research

Achievement Award, and was named “Engineering Manager of the Year” by the Engineering Management Society of the

IEEE. He is a fellow and member of the science council for the center for project and program management research at

NASA’s University Space Research Association. Shenhar holds five academic degrees including a BSc and MSc in

engineering from the Technion in Israel, MScs in statistics, and engineering economic systems, and PhD in engineering

from Stanford University. Professor Shenhar is the founder of the Technological Leadership Institute, a management

consulting company, which focuses on aligning business and technology.

DOV DVIR is the head of the Management Department at the Ben-Gurion University in Beer Sheba, Israel. Previously he

was the head of the Management of Technology (MOT) department at the Holon Center for Technological Education. He

holds a BS in electrical engineering from the Technion - Israel Institute of Technology, an MS in operations research and

MBA from Tel Aviv University, and a PhD in management (specialization in MOT) from Tel Aviv University. Professor Dvir

has accumulated more than 25 years in management of large-scale projects and consulting in the government and private

sectors. He has more than 80 publications in the areas of project management, technology management and technological

entrepreneurship, in refereed journals and scientfic conferences.


T omorrow’s projects should bebetter for our project experiencetoday. The keys to achieving this

simply stated but often elusive goalare lessons learned. Post-Project Reviewsto Gain Effective Lessons Learned byTerry Williams is a report on researchconducted at Strathclyde Universityand Southampton University, both inthe U.K.

The goal of the research was to identify current practicesand to compare them with advances in understanding projectbehavior to identify lessons that are not being learned.Research activities included a literature survey, a questionnaire,in-depth interviews in six selected organizations, and an analy-sis of a specific project.

An immediate and striking feature of this effort is thatalthough it is formal, academic research, it is immensely prac-tical. The research questions seem like things that might beasked and discussed around a project conference table, notmighty matters of inquiry reserved for the hallowed halls ofacademe. That may be the greatest contribution of thisresearch—it subjects to formal study important matters thatare often only anecdotally reviewed, or worse, taken for grant-ed and ignored.

The problem with lessons learned in a project environ-ment was first identified some 40 years ago. Projects are per-formed by groups of people who come together for theduration of the project and then go their own ways, never com-ing together again as the same group for the same or similarpurpose. Whatever was learned by the group through projectperformance is lost in the wind. Unless, of course, the organi-zation sponsoring the project makes some deliberate effort tocapture what was learned and make it available to future proj-ect groups.

Those efforts to capture learning are the subject of thisresearch’s literature survey.

A literature search, intended to survey current thinking ona specific subject, often provides a tutorial on that subject as aby-product. Williams’ search goes beyond this. It is truly a tourde force on lessons learned—something that, by itself, offersmuch value to project managers seeking a better tomorrow.

Williams also collected data on current practice, obtaining522 responses to a questionnaire distributed in the U.K., U.S.,and China. Results are displayed in a way that provides anoth-er stand-out feature of this report—the clarity of graphics.

Readers can grasp at a glance that which is described in detailin the text. Some results are not surprising. For example,mature project organizations are better at lessons learned thanthose in the beginning or mid-development stages of projectprocess implementation.

Some results may be more of a surprise. Learning net-works and communities of practice were considered importantby 89% of respondents, but only 12% were doing it. Lessonslearned databases were considered important by 87%, butonly 22% were doing it. The main reason for inaction was lackof employee time. This is rather like the old and not-so-funnyjoke about taking longer to cut trees because you “don’t havetime” to sharpen the saw. The research revealed other reasonsfor inaction: lack of management support, and lack of incen-tive, resources, and clear guidelines.

With all this data in hand, Williams conducted interviewswith project management staff in six organizations. This partof the report may be the most interesting to readers as theymay experience an “I’ve been there” reaction. The interviewswere semi-structured, allowing some flexibility in response.Responses appear to reflect a helpful candor. Participatingorganizations include a variety in both size and domain.

Williams closed his research by looking into a specificproject to see how lessons learned are gathered and dissemi-nated. This section may also be of high interest to readers. Thisapplication view gives “life” to what otherwise may stand asdisembodied data. Williams asks hard but relevant questionssuch as “Why?” and “So what?” He links the application to thedata from the previous literature survey, questionnaire, andinterviews. Readers will see the connection and may againexperience a “been there” response.

The report includes the questionnaire and results forthose inclined to review detail. More important, it closes witha section of concise study conclusions that summarize resultsand indicate possible next steps. Tomorrow’s projects shouldbe—must be—better because of today’s experience. Thisresearch report documents in a disciplined way why that is oris not, and how it can be better in the future.

Project Management Institute, 2007, ISBN 13:

9781933890241, ISBN 10: 193389024X, paperback, 162 pp.,

$29.95 Member, $39.95 Nonmember

Reviewed by Kenneth H. Rose, PMP, Director, Peninsula Center

for Project Management in Hampton, VA, USA, and winner of the

2006 PMI David I. Cleland Project Management Literature Award

POST-PROJECT REVIEWS TO GAINEFFECTIVE LESSONS LEARNEDBY TERRY WILLIAMS

COVER TO COVERKenneth H. Rose, PMP, Book Review Editor


PROJECT PORTFOLIO MANAGEMENT:TOOLS AND TECHNIQUESBY PARVIZ F. RAD AND GINGER LEVIN

A s an organization’s proj-ect management philoso-phies and methodologies

mature, there is a natural pro-gression to expand the projectand program management dis-ciplines to the enterprise level.This intersection of project/pro-gram management with theenterprise-level strategic busi-ness objectives is the realm ofproject portfolio management

(PPM). To many, this is uncharted ground. Project PortfolioManagement: Tools and Techniques by Parviz F. Rad andGinger Levin provides a helpful map.

Financial investors often have a collection or portfolioof individual and related group investments, with the ulti-mate focus on maximizing the performance (value) of theentire portfolio. This same strategy is why PPM is applied toour project portfolios: to help maximize the value of thetotal collection of projects and programs, and to ensure thatthe interactions among the various projects in the portfolioare synergistic in nature, not working at cross purposes tothe enterprise strategy.

Project portfolio management is a strategic, mission-driven approach that is concerned with the enterprise as awhole. The portfolio management lifecycle is much moreencompassing than the project or program managementlifecycle and involves “identifying enterprise opportunities,selecting the projects to help fulfill these opportunities,planning and executing these projects, and continuallyassessing the benefits of these projects to organizational suc-cess.” As with the project management lifecycle, continuousmonitoring of each project’s contribution and alignment tothe enterprise goals is key to the successful implementationof a PPM program.

Right from the start, Rad and Levine stress the inter-rela-tional and dynamic aspects of PPM. Success is no longermeasured solely by an individual project being completedon time and in budget, but rather by measuring that pro-ject’s contribution (value) to the enterprise as compared tothe contribution of other projects. There is also a higherdegree of focus on overall resource allocation, not necessar-ily in terms of maximizing individual utilization, but ratherin maximizing the total enterprise or portfolio value by hav-ing the right resources in the right place at the right time.This is a departure from the typical project staffing model.

In today’s highly competitive environment where weare all doing more with less, the evaluation of each project

at key stages is critical to help ensure that each individualproject is still in line with the enterprise strategy and that itis still adding the value that was expected from the originalinvestment. In the PPM approach, the project team is striv-ing to work on the edge of the “efficient frontier” curve,always trying to maximize the possible expected return forany given level of risk. This strategy also challenges the “risk-adverse” nature of some project teams, where that approachputs the project either inside the “efficient frontier” or lowon the return-on-investment side of the curve.

Rad and Levin also address the issue of cost to developand maintain a PPM program. They have identified twomajor cost components: implicit and explicit. From theimplicit point of view, if the enterprise has a lot of projectsthat are not aligned to the strategic objectives, then there isan inherent waste of money. On the explicit cost side, thereis the cost to set up and to operate the PPM process, as wellas the related team costs. The authors put forth some graph-ical data to illustrate that although the explicit cost of run-ning a PPM increases with maturity of the PPM operation,the total cost (implicit plus explicit) decreases as enterprisePPM maturity increases.

So what are the basic reasons for implementing a PPMprogram? Organizations that decide to implement this strat-egy are looking for increased operational efficiency,improved cost savings, and increased profits. As this strategyis at the enterprise level and not the individual project level,there are shifts in behavior that must be decided on in orderto implement these goals effectively. An organization mustdecide if the specific mission of the PPM program is that ofdata collection and reporting, or to have a broader, moreactive role in affecting the enterprise-level strategic planningand decision-making. The success of a PPM program isgauged by the enterprise being able to select the right projectsand abandon the inappropriate projects on a repeatablebasis, always improving the performance of the total invest-ment portfolio. PPM provides that mechanism through pro-cedures to facilitate a standardized approach to projectprioritization, including a formalized set of guidelines andspecific review gates to ensure strategic alignment to the busi-ness goals of the enterprise.

IIL Publishing, 2006, ISBN 13: 9780970827685, ISBN 10:

0970827687, paperback, 144 pp., $28.50 Member, $29.95

Nonmember

Reviewed by Greg Indelicato, PMP, Program/Project

Manager, EDS, Plano, TX, USA


THE PROGRAM MANAGEMENT OFFICE: ESTABLISHING,MANAGING AND GROWING THE VALUE OF A PMOBY CRAIG J. LETAVEC

T he Program ManagementOffice is a “how-to” bookwritten by someone with a

great deal of practical experience.Letavec gives the impression thathe is a battle-scarred veteran whonot only survived but also paidattention and learned from hisexperiences. His examples arerooted in reality and pragmatism.That experience enables him tooffer sound and valuable guid-ance for those considering imple-menting a program management

office (PMO) as well as those already working in a func-tioning PMO.

The book revolves around two core themes. First,Letavec describes the alternative roles of the PMO and theirattributes. Second, he describes how to help an organizationunderstand the benefits of a PMO and how to deploy one.

The first theme asserts a PMO may function in any ofthree roles:

• A consulting role—advising program teams on toolsand techniques for managing their programs

• A knowledge management role—manager of lessonslearned capture and dissemination

• A standards setting/compliance role—process standards,tools standards, reporting standards, etc. The early chap-ters outline this broad theme while the last three chap-ters describe each of the PMO roles in more detail.

The second theme addresses approaches and tech-niques for gaining organizational support for, and deploy-ing, a PMO to deliver the desired functionality. Letavecdescribes how one might characterize the benefits of a PMOin such a way that senior management can see the benefitsto the organization. He also describes in great detail aroadmap for deploying a PMO. The middle of the book isdedicated to this theme.

Letavec has partitioned the two themes in the text sothat it enables the reader to use the book more effectively asa reference tool. The reader is able to quickly find materialuseful in describing the PMO and making the argument forits implementation. Or, the reader can go directly to thematerial describing deployment methodology and chal-lenges—material that walks sequentially through theprocess from planning initial deployment to ongoing sus-tainance of the PMO.

The descriptions of the functions of the PMO are clearand relevant. The author also effectively addresses how toadapt the three fundamental roles to the pragmatics of aparticular organization. For example, he describes how anorganizational bias toward centralization versus decentral-ization may influence the structure of the PMO itself. Eachoperating business unit may have a somewhatautonomous PMO loosely aligned with a headquartersPMO or, on the other hand, the headquarters PMO mayprovide direct control across the business units. Letavecalso addresses the ideal level of resources versus what anorganization may be willing or able to commit for thePMO; that is, how to make progress quickly with limitedresources, thus building organizational support for addi-tional investment later.

He also describes how an organization’s cultural envi-ronment may affect the PMO. Employees not accustomedto standardization and process rigor may resist PMOattempts to establish and enforce standards. He alsodescribes how one must adapt the PMO function and struc-ture to existing organizational structures. The relationshipsbetween an existing program management department or aprogram administration department may influence whatfunctions are included within the PMO charter.

Latevec’s description of the basic structural model ofthe PMO is not without its minor flaws. He seems to havetrouble distinguishing, or at least articulating, the cus-tomer-facing role of the PMO from the customer-facingrole of the program manager. One figure depicts the PMOas the primary interface between the organization and thecustomer and some parts of the text reinforce that assertion.Yet, other parts of the text characterize the program manag-er as the primary customer interface. This apparent contra-diction might have been explained as a variation in thedeployment, but Latevec fails to make that point, leavingthe reader a bit confused.

But there is more to praise than criticize in this excellentbook. All in all, The Program Management Office is a soliddescription of the PMO, its benefits, and how it can be effec-tively deployed. The author knows the subject matter andcommunicates it effectively and well.

J. Ross Publishing, 2006, ISBN 13: 9781932159592, ISBN

10: 1932159592, hardcover, 408 pp., $59.95 list price

Reviewed by Jerry L. Wellman, PhD, PMP, Vice President,

Honeywell Technology Solutions, Inc., Columbia, MD, USA


THE POWER OF ENTERPRISE-WIDE PROJECT MANAGEMENTBY DENNIS L. BOLLES AND DARREL G. HUBBARD

T oday we have organizationsthat employ thousandsof Project Management

Professional (PMP®) credential hold-ers. We are talking here about com-panies such as IBM, EDS, Motorola,and Hewlett-Packard. These are largebureaucratic enterprises with manycorporate divisions and many layersof employees. Clearly, their commit-ment to employing project manage-

ment practices is substantial. A big challenge they face is howto incorporate project management into their operations sothat they can tap into the substantial project managementcapabilities that reside within their organizations.

The Power of Enterprise-Wide Project Management byDennis Bolles and Darrel Hubbard is directed at helpingsuch enterprises harness their project management capabili-ties. It is not enough for these organizations to train thou-sands of employees on good project management practiceand to support them to earn their PMP certification. Theymust also figure out how to adjust their organizations to useall this hard-earned talent effectively. Bolles and Hubbard’sbook aims to help them in this effort.

Enterprise-wide project management is based on estab-lishing an enterprise-wide project office (EPMO) whosefunction is to lead the project management efforts of theorganization. Because we are dealing with huge enterpriseshere, Bolles and Hubbard propose a model that employsmultiple levels of project offices. At the pinnacle, we have theEPMO, which oversees all project management policy withinthe organization. At the next level, we encounter theDivisional Project Management Office (DPMO), whichfocuses on project management efforts at the divisional level.Next, we have the Business Unit Project Management Office(BUPMO) and then the basic Project Management Office(PMO) that actually addresses the execution of real projects.

This book is a prescriptive “how-to” book that lays outthe steps that large organizations should follow to establishenterprise-wide project management policies and practices. Itis filled with bulleted checklists of things that should be takeninto account when implementing mature project manage-ment practices in organizations. It contains page after page oftables that show you how to implement enterprise-wide proj-ect management, based on the Project Management Institute’sfive project management process groups: project initiation,project planning, project execution, project monitoring andcontrol, and project closure. If I were charged with imple-menting good project management practice at HumongousInc., I would certainly study this book carefully. Its prescrip-tion is realistic and accurately reflects conditions faced bylarge, bureaucratic business enterprises.

PMI project management geeks beware! The book hard-ly looks at project management tools, and with good reason.It takes the position that when implementing project man-agement practices in large organizations, we need toapproach the effort as a business undertaking. PERT/CPM,Gantt charts, and cost curves are secondary (perhaps eventertiary) concerns. The real issue is: How to implement proj-ect management to serve the organization’s strategic busi-ness goals. This makes a lot of sense to me. When dealingwith the affairs of multibillion-dollar enterprises, who caresif you standardize on Microsoft Project or Primavera? Theimportant issue is: Are we implementing project manage-ment practices that enable the enterprise to achieve itsstrategic business goals?

This is a thoughtful book, based on the substantial proj-ect management experiences of the authors. As someonewho has been working with project offices since the late1980s, many of the observations made by Bolles andHubbard resonated with me. Like the authors, I too wasamazed in the early 2000s to see large enterprises obliteratetheir project management initiatives with the downturn ofthe economy. I too recognize that unless project manage-ment is institutionalized in enterprises, it can be cruciallyimportant to the enterprise one day, then gone the next.Also, I think that the checklists and tables that abound inthe book are useful to anyone wishing to establish coherentproject management capabilities in large organizations.

My principal criticism of the book is that it shouldillustrate its points with more concrete examples. For exam-ple, it dedicates a chapter to the use of work breakdownstructures (WBSs) and suggests that WBSs can be employedto lay out the operations of the enterprise-wide projectoffice. But no sample WBS is offered and I can only guess athow a WBS can be employed in the way the authors sug-gest. Another example: the book states that enterprise-wideproject management is being adopted by project-basedorganizations, but it doesn’t name names. Exactly whatorganizations have adopted enterprise-wide project man-agement practices? To what extent do the adopted practicesin these organizations reflect the methodology that Bollesand Hubbard propose?

Still, this insightful and realistic book provides a lot offood for thought and practical advice for anyone thinkingabout establishing an EPMO at Humongous Inc.

Amacom, 2007, ISBN 13: 9780814474044, ISBN 10:

0814474047, hardcover, 343 pp., $47.45 Members, $49.95

Nonmembers

Reviewed by J. Davidson Frame, PhD, PMP, Academic Dean,

University of Management and Technology, Arlington,

Virginia, USA


CALENDAR OF EVENTS

16-19 July 2007PMI MEGA SeminarsWorld® Orlando, Florida, USA. For more infor-mation, please visit the SeminarsWorld Homepage

19 July – 24 AugusteSeminarsWorldSM Course: Project ManagementFundamentals: An Intensive Program. Get up to speed in Project Management without leaving your office through aneSeminarsWorld course. For more information, please visit theeSeminarsWorld Homepage

26 July – 9 August eSeminarsWorldSM Course: Managing Multiple Projects.Gain extensive understanding of how risk, human resources,schedules and cost must be integrated. For more information,please visit the eSeminarsWorld Homepage

AUGUST 20079-23 August eSeminarsWorldSM Course: Building Teams, Commitment andCulture for Virtual Teams. Benefit from semi-synchronous learningmethodology, expert instructors and classmates from around the world.For more information, please visit the eSeminarsWorld Homepage

13-16 August PMI SeminarsWorld® Washington, D.C., USA. For more information, please visit the SeminarsWorld Homepage

16-30 August eSeminarsWorldSM Course: Project Management for theExperienced Professional. This intensive program provides participants with practical experience applying the fundamentals of Project Management. For more information, please visit theeSeminarsWorld Homepage

25 AugustPMI Atlanta Chapter Annual Profession Development Day. Atlanta,Georgia, USA. For more information, please visit www.pmiatlanta.org

28-31 August4th annual PMOz project management Australia Conference.Gold Coast, Queensland, Australia. The conference, themed “Creating Success,” will be held in conjunction with Bond Universityand co-located with SEPG conference Keynote speakers include Ed Hoffman of NASA, Peter Head of Ove Arup and Bernard Salt ofKPMG. For more information, please visit www.pmoz.com.au

JUNE 20072-5 June PMI Information Systems SIG’s PDS’07 Symposium &Workshops. Scottsdale, Arizona, USA. IS/IT specific hands-on workshops, interesting and informative speakers, sponsor presentationsand networking. For more information, please visit pmi-issig.org/pds

7-21 June eSeminarsWorldSM Course: Effective Project Communicationand Control for Virtual Teams. This course addresses how project managers can set up an infrastructure which facilitates projectcommunication and control. For more information, please visit theeSeminarsWorld Homepage

11-14 June PMI SeminarsWorld® San Jose, California, USA. For more information, please visit the SeminarsWorld Homepage

18-21 June Project & Portfolio Management Event, sponsored byGartner. Boston, Massachusetts, USA. Now in its fifth year, Project & Portfolio Management is a premier global event for enterprise PPM.It showcases the latest applications for optimizing business processesaround the prioritization and development of both new products andinformation systems in project-intensive environments. Presentationsare in three tracks: IT; Product Development; and Government.Technology providers will provide Case Study Presentations. Generalsessions and one-on-ones will be conducted by over a dozen Gartner thought leaders. For more information, please visit www.ppmsummit.com

19-22 JuneProjectWorld & the World Congress for Business AnalystsRegional. Boston, Massachusetts, USA. For more information, pleasevisit www.projectworldregional.com

JULY 200712-26 July eSeminarsWorldSM Course: Program Management in Context:Developing Business Strategies Through Programs. Learn systematic management of projects through programs to achievebusiness strategy and deliver business benefits. For more information,please visit the eSeminarsWorld Homepage

104

Case Studies in Project Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .C2

Program and Portfolio Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .C3

Call for Papers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .C4

Index of Advertisers

PROJECT

PROGRAM

PORTFOLIO

MANAGE SUCCESS.Organizations invest in their future by initiating projects, programs and

portfolios as a means of achieving organizational success. The utilization

of program and portfolio management processes and skilled program and

portfolio managers continues to rise. Recognizing this development, PMI

introduces The Standard for Program Management and The Standard

for Portfolio Management.

Learn how to optimize program components and organizational schedules and resources for success. The Standard for Program Management describes the processes and provides guidance for managing multiple projects and non-project activities within a program environment.

PMI Member Price : $38.95 (US)Price : $46.95 (US)

Portfolio management is gaining favor as a means of helping organizations focus not only on “doing work right” but also on “doing the right work.” The Standard for Portfolio Management provides methods to take a comprehensive view of portfolios, group them for the most effective management and ensure all components are aligned strategically.

PMI Member Price : $38.95 (US)Price : $46.95 (US)

Making project management indispensable for business results.

Take the next step toward managing the success of your organization. Visit the PMI Bookstore (www.pmibookstore.org) for more information and to order your copy today.

Global Operations CenterFour Campus BoulevardNewtown Square, PA 19073-3299 USATel: +1-610-356-4600Fax: +1-610-356-4647E-mail: [email protected]: www.pmi.org

©2006 Project Management Institute, Inc. All rights reserved.“PMI”, the PMI logo and “Making project management indispensable for business results” are marks of Project Management Institute, Inc.

045-009-2006 (03-06)

®

Regional Service CentreEurope-Middle East-Africa (EMEA)Avenue de Tervueren 300B-1150 Brussels, BelgiumTel: +32-2-743 15 73Fax: +32-2-743 15 50E-mail: [email protected]

Regional Service CentreAsia Pacific73 Bukit Timah Road#03-01 Rex HouseSingapore, 229832Tel: +65 6330 6733Fax: +65 6336 2263E-mail: [email protected]

project management ournal · tel: +1-610-356-4600, ... project management institute global...

Documents