harvard business review on turnarounds

HarvardBusinessReview

HBR029chfm 8/15/01 1:32 PM Page i

The series is designed to bring today’s managers and professionals thefundamental information they need to stay competitive in a fast-moving world. From the preeminent thinkers whose work has definedan entire field to the rising stars who will redefine the way we thinkabout business, here are the leading minds and landmark ideas thathave established the Harvard Business Review as required reading forambitious businesspeople in organizations around the globe.

Other books in the series:

Harvard Business Review Interviews with CEOs

Harvard Business Review on Brand Management

Harvard Business Review on Breakthrough Thinking

Harvard Business Review on Business and the Environment

Harvard Business Review on the Business Value of IT

Harvard Business Review on Change

Harvard Business Review on Corporate Governance

Harvard Business Review on Corporate Strategy

Harvard Business Review on Crisis Management

Harvard Business Review on Decision Making

Harvard Business Review on Effective Communication

Harvard Business Review on Entrepreneurship

Harvard Business Review on Finding and Keeping the Best People

Harvard Business Review on Innovation

Harvard Business Review on Knowledge Management

Harvard Business Review on Leadership

Harvard Business Review on Managing High-Tech Industries

Harvard Business Review on Managing People

Harvard Business Review on Managing Uncertainty

HBR029chfm 8/15/01 1:32 PM Page ii

Other books in the series (continued):

Harvard Business Review on Managing the Value Chain

Harvard Business Review on Measuring Corporate Performance

Harvard Business Review on Mergers and Acquisitions

Harvard Business Review on Negotiation and Conflict Resolution

Harvard Business Review on Nonprofits

Harvard Business Review on Organizational Learning

Harvard Business Review on Strategies for Growth

Harvard Business Review on What Makes a Leader

Harvard Business Review on Work and Life Balance

HBR029chfm 8/15/01 1:32 PM Page iii

HBR029chfm 8/15/01 1:32 PM Page iv

This Page Intentionally Left Blank


HBR029chfm 8/15/01 1:32 PM Page v

Copyright 1997, 1999, 2000, 2001Harvard Business School Publishing CorporationAll rights reservedPrinted in the United States of America05 04 03 02 01 5 4 3 2 1

All rights reserved. No part of this book may be reproduced, stored in aretrieval system, or transmitted, in any form or by any means, elec-tronic, mechanical, photocopying, recording, or otherwise without theprior written permission of the copyright holder.

The Harvard Business Review articles in this collection are available asindividual reprints. Discounts apply to quantity purchases. For informa-tion and ordering, please contact Customer Service, Harvard BusinessSchool Publishing, Boston, MA 02163. Telephone: (617) 783-7500 or (800) 988-0886, 8 A.M. to 6 P.M. Eastern Time, Monday through Friday.Fax: (617) 783-7555, 24 hours a day. E-mail: [email protected]

Library of Congress Cataloging-in-Publication DataHarvard business review on turnarounds.

p. cm. — (A Harvard business review paperback)Includes bibliographical references and index.ISBN 1-57851-636-6 (alk. paper)1. Corporate turnarounds. I. Title: Turnarounds. II. Harvard

business review. III. Harvard business review series.HD58.8 .H3696 2001658.4´063—dc21 2001039413

CIP

The paper used in this publication meets the requirements of the Ameri-can National Standard for Permanence of Paper for Publications andDocuments in Libraries and Archives Z39.48-1992.

HBR029chfm 8/15/01 1:32 PM Page vi

Cracking the Code of Change 1

Turning Goals into Results:The Power of Catalytic Mechanisms 23

Changing the Way We Change 55 , ,

Racing for Growth:An Interview with PerkinElmer’s Greg Summe 89

The Tough Work of Turning Around a Team 105

Saving Money, Saving Lives 115

Harley’s Leadership U-Turn 133

Waking Up IBM:How a Gang of Unlikely Rebels Transformed Big Blue 145

About the Contributors 167

Index 173

Contents

vii

HBR029chfm 8/15/01 1:32 PM Page vii

HBR029chfm 8/15/01 1:32 PM Page viii



HBR029chfm 8/15/01 1:32 PM Page ix

HBR029chfm 8/15/01 1:32 PM Page viii


Cracking the Code of Change

Executive Summary

TODAY’S FAST-PACED ECONOMY demands that busi-nesses change or die. But few companies manage cor-porate transformations as well as they would like. Thebrutal fact is that about 70% of all change initiatives fail.

In this article, authors Michael Beer and Nitin Nohriadescribe two archetypes—or theories—of corporate trans-formation that may help executives crack the code ofchange. Theory E is change based on economic value:shareholder value is the only legitimate measure of suc-cess, and change often involves heavy use of economicincentives, layoffs, downsizing, and restructuring. TheoryO is change based on organizational capability: thegoal is to build and strengthen corporate culture.

Most companies focus purely on one theory or theother, or haphazardly use a mix of both, the authors say. Combining E and O is directionally correct, they

1

HBR029ch1 8/15/01 1:34 PM Page 1

contend, but it requires a careful, conscious integrationplan. Beer and Nohria present the examples of two com-panies, Scott Paper and Champion International, thatused a purely E or purely O strategy to create change—and met with limited levels of success.

They contrast those corporate transformations withthat of UK-based retailer ASDA, which has successfullyembraced the paradox between the opposing theoriesof change and integrated E and O. The lesson fromASDA? To thrive and adapt in the new economy, com-panies must make sure the E and O theories of businesschange are in sync at their own organizations.

T has ushered in great businessopportunities—and great turmoil. Not since the Indus-trial Revolution have the stakes of dealing with changebeen so high. Most traditional organizations haveaccepted, in theory at least, that they must either changeor die. And even Internet companies such as eBay, Ama-zon.com, and America Online recognize that they needto manage the changes associated with rapidentrepreneurial growth. Despite some individual suc-cesses, however, change remains difficult to pull off, andfew companies manage the process as well as they wouldlike. Most of their initiatives—installing new technology,downsizing, restructuring, or trying to change corporateculture—have had low success rates. The brutal fact isthat about 70% of all change initiatives fail.

In our experience, the reason for most of those fail-ures is that in their rush to change their organizations,managers end up immersing themselves in an alphabetsoup of initiatives. They lose focus and become mesmer-

2 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 2

ized by all the advice available in print and on-line aboutwhy companies should change, what they should try toaccomplish, and how they should do it. This proliferation

of recommendationsoften leads to muddlewhen change isattempted. The resultis that most changeefforts exert a heavytoll, both human andeconomic. To improvethe odds of success,

and to reduce the human carnage, it is imperative thatexecutives understand the nature and process of corpo-rate change much better. But even that is not enough.Leaders need to crack the code of change.

For more than 40 years now, we’ve been studying thenature of corporate change. And although every busi-ness’s change initiative is unique, our research suggeststhere are two archetypes, or theories, of change. Thesearchetypes are based on very different and often uncon-scious assumptions by senior executives—and the con-sultants and academics who advise them—about whyand how changes should be made. Theory E is changebased on economic value. Theory O is change based onorganizational capability. Both are valid models; eachtheory of change achieves some of management’s goals,either explicitly or implicitly. But each theory also has itscosts—often unexpected ones.

Theory E change strategies are the ones that makeall the headlines. In this “hard” approach to change,shareholder value is the only legitimate measure ofcorporate success. Change usually involves heavy use ofeconomic incentives, drastic layoffs, downsizing, and

Theory E change strategiesusually involve heavy useof economic incentives,drastic layoffs, downsizing,and restructuring. Shareholdervalue is the only legitimatemeasure of corporate success.


HBR029ch1 8/15/01 1:34 PM Page 3

restructuring. E change strategies are more commonthan O change strategies among companies in theUnited States, where financial markets push corporateboards for rapid turnarounds. For instance, whenWilliam A. Anders was brought in as CEO of GeneralDynamics in 1991, his goal was to maximize economicvalue—however painful the remedies might be. Overthe next three years, Anders reduced the workforce by71,000 people—44,000 through the divestiture of sevenbusinesses and 27,000 through layoffs and attrition.Anders employed common E strategies.

Managers who subscribe to Theory O believe that ifthey were to focus exclusively on the price of their stock,they might harm their organizations. In this “soft”

approach to change,the goal is to developcorporate culture andhuman capabilitythrough individual andorganizational learn-ing—the process ofchanging, obtainingfeedback, reflecting,and making furtherchanges. U.S. compa-

nies that adopt O strategies, as Hewlett-Packard didwhen its performance flagged in the 1980s, typically havestrong, long-held, commitment-based psychological con-tracts with their employees.

Managers at these companies are likely to see therisks in breaking those contracts. Because they place ahigh value on employee commitment, Asian and Euro-pean businesses are also more likely to adopt an O strat-egy to change.

Theory O changestrategies are geared toward building up the corporateculture: employee behaviors,attitudes, capabilities, andcommitment. The organiza-tion’s ability to learn from itsexperiences is a legitimateyardstick of corporate success.

4 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 4

Few companies subscribe to just one theory. Mostcompanies we have studied have used a mix of both. Butall too often, managers try to apply theories E and O intandem without resolving the inherent tensions betweenthem. This impulse to combine the strategies is direc-tionally correct, but theories E and O are so different thatit’s hard to manage them simultaneously—employeesdistrust leaders who alternate between nurturing andcutthroat corporate behavior. Our research suggests,however, that there is a way to resolve the tension so thatbusinesses can satisfy their shareholders while buildingviable institutions. Companies that effectively combinehard and soft approaches to change can reap big payoffsin profitability and productivity. Those companies aremore likely to achieve a sustainable competitive advan-tage. They can also reduce the anxiety that grips wholesocieties in the face of corporate restructuring.

In this article, we will explore how one company suc-cessfully resolved the tensions between E and O strate-gies. But before we do that, we need to look at just howdifferent the two theories are.

A Tale of Two Theories

To understand how sharply theories E and O differ, wecan compare them along several key dimensions ofcorporate change: goals, leadership, focus, process,reward system, and use of consultants. (For a side-by-side comparison, see the exhibit “Comparing Theoriesof Change.”) We’ll look at two companies in similarbusinesses that adopted almost pure forms of eacharchetype. Scott Paper successfully used Theory E toenhance shareholder value, while Champion Interna-tional used Theory O to achieve a complete cultural


HBR029ch1 8/15/01 1:34 PM Page 5

transformation that increased its productivity andemployee commitment. But as we will soon observe,both paper producers also discovered the limitations ofsticking with only one theory of change. Let’s comparethe two companies’ initiatives.

6 Beer and Nohria

Comparing Theories of Change

Our research has shown that all corporate transformations can becompared along the six dimensions shown here. The table outlines thedifferences between the E and O archetypes and illustrates what an inte-grated approach might look like.

Dimensionsof Change

Goals

Leadership

Focus

Process

Reward System

Use ofConsultants

Theory E

Maximize share-holder value

Manage changefrom the top down

Emphasize struc-ture and systems

Plan and establishprograms

Motivate throughfinancial incen-tives

Consultants ana-lyze problems andshape solutions

Theory O

Develop organiza-tional capabilities

Encourage partici-pation from thebottom up

Build up corporateculture: employees’behavior and atti-tudes

Experiment andevolve

Motivate throughcommitment—use pay as fairexchange

Consultants sup-port managementin shaping theirown solutions

Theories E andO Combined

Explicitly embracethe paradoxbetween economicvalue and organiza-tional capability

Set direction fromthe top and engagethe people below

Focus simultane-ously on the hard(structures and sys-tems) and the soft(corporate culture)

Plan for spontaneity

Use incentives toreinforce changebut not to drive it

Consultants areexpert resourceswho empoweremployees

HBR029ch1 8/15/01 1:34 PM Page 6

When Al Dunlap assumed leadership of Scott Paper inMay 1994, he immediately fired 11,000 employees andsold off several businesses. His determination to restruc-ture the beleaguered company was almost monomania-cal. As he said in one of his speeches: “Shareholders arethe number one constituency. Show me an annual reportthat lists six or seven constituencies, and I’ll show you amismanaged company.” From a shareholder’s perspec-tive, the results of Dunlap’s actions were stunning. Injust 20 months, he managed to triple shareholder returnsas Scott Paper’s market value rose from about $3 billionin 1994 to about $9 billion by the end of 1995. The finan-cial community applauded his efforts and hailed ScottPaper’s approach to change as a model for improvingshareholder returns.

Champion’s reform effort couldn’t have been more dif-ferent. CEO Andrew Sigler acknowledged that enhancedeconomic value was an appropriate target for manage-ment, but he believed that goal would be best achieved bytransforming the behaviors of management, unions, and workers alike. In 1981, Sigler and other managerslaunched a long-term effort to restructure corporate cul-ture around a new vision called the Champion Way, a setof values and principles designed to build up the compe-tencies of the workforce. By improving the organization’scapabilities in areas such as teamwork and communica-tion, Sigler believed he could best increase employee pro-ductivity and thereby improve the bottom line.

Leaders who subscribe to Theory E manage change theold-fashioned way: from the top down. They set goals


HBR029ch1 8/15/01 1:34 PM Page 7

with little involvement from their management teamsand certainly without input from lower levels or unions.Dunlap was clearly the commander in chief at ScottPaper. The executives who survived his purges, for exam-ple, had to agree with his philosophy that shareholdervalue was now the company’s primary objective. Nothingmade clear Dunlap’s leadership style better than thenickname he gloried in: “Chainsaw Al.”

By contrast, participation (a Theory O trait) was thehallmark of change at Champion. Every effort was madeto get all its employees emotionally committed toimproving the company’s performance. Teams draftedvalue statements, and even the industry’s unions werebrought into the dialogue. Employees were encouragedto identify and solve problems themselves. Change atChampion sprouted from the bottom up.

In E-type change, leaders typically focus immediately onstreamlining the “hardware” of the organization—thestructures and systems. These are the elements that canmost easily be changed from the top down, yielding swiftfinancial results. For instance, Dunlap quickly decided tooutsource many of Scott Paper’s corporate functions—benefits and payroll administration, almost all of itsmanagement information systems, some of its technol-ogy research, medical services, telemarketing, and secu-rity functions. An executive manager of a global mergerexplained the E rationale: “I have a [profit] goal of $176million this year, and there’s no time to involve others ordevelop organizational capability.”

By contrast, Theory O’s initial focus is on building upthe “software” of an organization—the culture, behavior,and attitudes of employees. Throughout a decade of

8 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 8

reforms, no employees were laid off at Champion. Rather,managers and employees were encouraged to collectivelyreexamine their work practices and behaviors with a goalof increasing productivity and quality. Managers werereplaced if they did not conform to the new philosophy,but the overall firing freeze helped to create a culture oftrust and commitment. Structural change followed oncethe culture changed. Indeed, by the mid-1990s, Championhad completely reorganized all its corporate functions.Once a hierarchical, functionally organized company,Champion adopted a matrix structure that empoweredemployee teams to focus more on customers.

Theory E is predicated on the view that no battle can bewon without a clear, comprehensive, common plan ofaction that encourages internal coordination andinspires confidence among customers, suppliers, andinvestors. The plan lets leaders quickly motivate andmobilize their businesses; it compels them to take tough,decisive actions they presumably haven’t taken in thepast. The changes at Scott Paper unfolded like a militarybattle plan. Managers were instructed to achieve specifictargets by specific dates. If they didn’t adhere to Dunlap’stightly choreographed marching orders, they riskedbeing fired.

Meanwhile, the changes at Champion were moreevolutionary and emergent than planned and program-matic. When the company’s decade-long reform began in1981, there was no master blueprint. The idea was thatinnovative work processes, values, and culture changesin one plant would be adapted and used by other plantson their way through the corporate system. No singleperson, not even Sigler, was seen as the driver of change.


HBR029ch1 8/15/01 1:34 PM Page 9

Instead, local leaders took responsibility. Top manage-ment simply encouraged experimentation from theground up, spread new ideas to other workers, and trans-ferred managers of innovative units to lagging ones.

The rewards for managers in E-type change programs areprimarily financial. Employee compensation, for example,is linked with financial incentives, mainly stock options.Dunlap’s own compensation package—which ultimatelynetted him more than $100 million—was tightly linked toshareholders’ interests. Proponents of this system arguethat financial incentives guarantee that employees’ inter-ests match stockholders’ interests. Financial rewards alsohelp top executives feel compensated for a difficult job—one in which they are often reviled by their onetime col-leagues and the larger community.

The O-style compensation systems at Champion rein-forced the goals of culture change, but they didn’t drivethose goals. A skills-based pay system and a corpo-ratewide gains-sharing plan were installed to draw unionworkers and management into a community of purpose.Financial incentives were used only as a supplement tothose systems and not to push particular reforms. WhileChampion did offer a companywide bonus to achievebusiness goals in two separate years, this came late in thechange process and played a minor role in actually fulfill-ing those goals.

Theory E change strategies often rely heavily on externalconsultants. A SWAT team of Ivy League–educated

10 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 10

MBAs, armed with an arsenal of state-of-the-art ideas, isbrought in to find new ways to look at the business andmanage it. The consultants can help CEOs get a fix onurgent issues and priorities. They also offer much-neededpolitical and psychological support for CEOs who areunder fire from financial markets. At Scott Paper, Dunlapengaged consultants to identify many of the painful cost-savings initiatives that he subsequently implemented.

Theory O change programs rely far less on consul-tants. The handful of consultants who were introducedat Champion helped managers and workers make their

own business analysesand craft their ownsolutions. And whilethe consultants hadtheir own ideas, theydid not recommendany corporate program,dictate any solutions,or whip anyone intoline. They simply led a

process of discovery and learning that was intended tochange the corporate culture in a way that could not beforeseen at the outset.

In their purest forms, both change theories clearlyhave their limitations. CEOs who must make difficult E-style choices understandably distance themselves fromtheir employees to ease their own pain and guilt. Onceremoved from their people, these CEOs begin to see theiremployees as part of the problem. As time goes on, theseleaders become less and less inclined to adopt O-stylechange strategies. They fail to invest in building thecompany’s human resources, which inevitably hollowsout the company and saps its capacity for sustained

Scott Paper’s CEOtrebled shareholder returnsbut failed to buildthe capabilities needed forsustained competitiveadvantage—commitment,coordination, communication,and creativity.


HBR029ch1 8/15/01 1:34 PM Page 11

performance. At Scott Paper, for example, Dunlap tre-bled shareholder returns but failed to build the capabili-ties needed for sustained competitive advantage—com-mitment, coordination, communication, and creativity.In 1995, Dunlap sold Scott Paper to its longtime com-petitor Kimberly-Clark.

CEOs who embrace Theory O find that their loyaltyand commitment to their employees can prevent themfrom making tough decisions. The temptation is to post-pone the bitter medicine in the hopes that rising produc-tivity will improve the business situation. But productiv-ity gains aren’t enough when fundamental structural

change is required. Thatreality is underscored bytoday’s global financialsystem, which makescorporate performanceinstantly transparent tolarge institutional share-holders whose fund man-

agers are under enormous pressure to show good results.Consider Champion. By 1997, it had become one of theleaders in its industry based on most performance mea-sures. Still, newly instated CEO Richard Olsen was forcedto admit a tough reality: Champion shareholders had notseen a significant increase in the economic value of thecompany in more than a decade. Indeed, when Cham-pion was sold recently to Finland-based UPM-Kymmene,it was acquired for a mere 1.5 times its original sharevalue.

Managing the Contradictions

Clearly, if the objective is to build a company that canadapt, survive, and prosper over the years, Theory E

CEOs who embrace Theory Ofind that their loyaltyand commitment to theiremployees can preventthem from making toughdecisions.

12 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 12

strategies must somehow be combined with Theory Ostrategies. But unless they’re carefully handled, meldingE and O is likely to bring the worst of both theories andthe benefits of neither. Indeed, the corporate changeswe’ve studied that arbitrarily and haphazardly mixed Eand O techniques proved destabilizing to the organiza-tions in which they were imposed. Managers in thosecompanies would certainly have been better off to pickeither pure E or pure O strategies—with all their costs.At least one set of stakeholders would have benefited.

The obvious way to combine E and O is to sequencethem. Some companies, notably General Electric, havedone this quite successfully. At GE, CEO Jack Welchbegan his sequenced change by imposing an E-typerestructuring. He demanded that all GE businesses befirst or second in their industries. Any unit that failedthat test would be fixed, sold off, or closed. Welch fol-lowed that up with a massive downsizing of the GEbureaucracy. Between 1981 and 1985, total employmentat the corporation dropped from 412,000 to 299,000.Sixty percent of the corporate staff, mostly in planningand finance, was laid off. In this phase, GE people beganto call Welch “Neutron Jack,” after the fabled bomb thatwas designed to destroy people but leave buildingsintact. Once he had wrung out the redundancies, how-ever, Welch adopted an O strategy. In 1985, he started aseries of organizational initiatives to change GE culture.He declared that the company had to become “bound-aryless,” and unit leaders across the corporation had tosubmit to being challenged by their subordinates in openforum. Feedback and open communication eventuallyeroded the hierarchy. Soon Welch applied the new orderto GE’s global businesses.

Unfortunately for companies like Champion,sequenced change is far easier if you begin, as Welch did,


HBR029ch1 8/15/01 1:34 PM Page 13

with Theory E. Indeed, it is highly unlikely that E wouldsuccessfully follow O because of the sense of betrayalthat would involve. It is hard to imagine how a draconianprogram of layoffs and downsizing can leave intact thepsychological contract and culture a company has so

patiently built up overthe years. But whateverthe order, one sure prob-lem with sequencing isthat it can take a verylong time; at GE it hastaken almost twodecades. A sequencedchange may also requiretwo CEOs, carefully cho-

sen for their contrasting styles and philosophies, whichmay create its own set of problems. Most turnaroundmanagers don’t survive restructuring—partly because oftheir own inflexibility and partly because they can’t livedown the distrust that their ruthlessness has earnedthem. In most cases, even the best-intentioned effort torebuild trust and commitment rarely overcomes a bloodypast. Welch is the exception that proves the rule.

So what should you do? How can you achieve rapidimprovements in economic value while simultaneouslydeveloping an open, trusting corporate culture? Paradox-ical as those goals may appear, our research shows that itis possible to apply theories E and O together. It requiresgreat will, skill—and wisdom. But precisely because it ismore difficult than mere sequencing, the simultaneoususe of O and E strategies is more likely to be a source ofsustainable competitive advantage.

One company that exemplifies the reconciliation ofthe hard and soft approaches is ASDA, the UK grocery

To thrive and adapt in the new economy, companies must simultaneously build uptheir corporate cultures and enhance shareholder value; the O and E theories of business change must be in perfect step.

14 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 14

chain that CEO Archie Norman took over in December1991, when the retailer was nearly bankrupt. Normanlaid off employees, flattened the organization, and soldoff losing businesses—acts that usually spawn distrustamong employees and distance executives from theirpeople. Yet during Norman’s eight-year tenure as CEO,ASDA also became famous for its atmosphere of trustand openness. It has been described by executives atWal-Mart—itself famous for its corporate culture—asbeing “more like Wal-Mart than we are.” Let’s look athow ASDA resolved the conflicts of E and O along the sixmain dimensions of change.

Explicitly confront the tension between E and Ogoals. With his opening speech to ASDA’s executiveteam—none of whom he had met—Norman indicatedclearly that he intended to apply both E and O strategiesin his change effort. It is doubtful that any of his listenersfully understood him at the time, but it was importantthat he had no conflicts about recognizing the paradoxbetween the two strategies for change. He said as muchin his maiden speech: “Our number one objective is tosecure value for our shareholders and secure the tradingfuture of the business. I am not coming in with any magi-cal solutions. I intend to spend the next few weeks listen-ing and forming ideas for our precise direction. . . . Weneed a culture built around common ideas and goals thatinclude listening, learning, and speed of response, fromthe stores upwards. [But] there will be managementreorganization. My objective is to establish a clear focuson the stores, shorten lines of communication, and buildone team.” If there is a contradiction between building ahigh-involvement organization and restructuring toenhance shareholder value, Norman embraced it.


HBR029ch1 8/15/01 1:34 PM Page 15

Set direction from the top and engage peoplebelow. From day one, Norman set strategy withoutexpecting any participation from below. He said ASDAwould adopt an everyday-low-pricing strategy, and Nor-man unilaterally determined that change would begin byhaving two experimental store formats up and runningwithin six months. He decided to shift power from theheadquarters to the stores, declaring: “I want everyone tobe close to the stores. We must love the stores to death;that is our business.” But even from the start, there wasan O quality to Norman’s leadership style. As he put it inhis first speech: “First, I am forthright, and I like to argue.Second, I want to discuss issues as colleagues. I am look-ing for your advice and your disagreement.” Normanencouraged dialogue with employees and customersthrough colleague and customer circles. He set up a “TellArchie” program so that people could voice their con-cerns and ideas.

Making way for opposite leadership styles was also anessential ingredient to Norman’s—and ASDA’s—success.This was most clear in Norman’s willingness to hireAllan Leighton shortly after he took over. Leighton even-tually became deputy chief executive. Norman andLeighton shared the same E and O values, but they hadcompletely different personalities and styles. Norman,cool and reserved, impressed people with the power ofhis mind—his intelligence and business acumen.Leighton, who is warmer and more people oriented,worked on employees’ emotions with the power of hispersonality. As one employee told us, “People respectArchie, but they love Allan.” Norman was the first tocredit Leighton with having helped to create emotionalcommitment to the new ASDA. While it might be possi-ble for a single individual to embrace opposite leadership

16 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 16

styles, accepting an equal partner with a very differentpersonality makes it easier to capitalize on those styles.Leighton certainly helped Norman reach out to the orga-nization. Together they held quarterly meetings withstore managers to hear their ideas, and they supple-mented those meetings with impromptu talks.

Focus simultaneously on the hard and soft sides ofthe organization. Norman’s immediate actions fol-lowed both the E goal of increasing economic value andthe O goal of transforming culture. On the E side, Nor-man focused on structure. He removed layers of hierar-chy at the top of the organization, fired the financial offi-cer who had been part of ASDA’s disastrous policies, anddecreed a wage freeze for everyone—management andworkers alike. But from the start, the O strategy was anequal part of Norman’s plan. He bought time for all thischange by warning the markets that financial recoverywould take three years. Norman later said that he spent75% of his early months at ASDA as the company’shuman resource director, making the organization lesshierarchical, more egalitarian, and more transparent.Both Norman and Leighton were keenly aware that theyhad to win hearts and minds. As Norman put it to work-ers: “We need to make ASDA a great place for everyoneto work.”

Plan for spontaneity. Training programs, total-qualityprograms, and top-driven culture change programsplayed little part in ASDA’s transformation. From thestart, the ASDA change effort was set up to encourageexperimentation and evolution. To promote learning, forexample, ASDA set up an experimental store that waslater expanded to three stores. It was declared a risk-free


HBR029ch1 8/15/01 1:34 PM Page 17

zone, meaning there would be no penalties for failure. Across-functional task force “renewed,” or redesigned,ASDA’s entire retail proposition, its organization, and itsmanagerial structure. Store managers were encouragedto experiment with store layout, employee roles, rangesof products offered, and so on. The experiments pro-duced significant innovations in all aspects of store oper-ations. ASDA’s managers learned, for example, that theycouldn’t renew a store unless that store’s managementteam was ready for new ideas. This led to an innovationcalled the Driving Test, which assessed whether storemanagers’ skills in leading the change process werealigned with the intended changes. The test perfectlyillustrates how E and O can come together: it bubbled upO-style from the bottom of the company, yet it boundmanagers in an E-type contract. Managers who failed thetest were replaced.

Let incentives reinforce change, not drive it. Anysynthesis of E and O must recognize that compensationis a double-edged sword. Money can focus and motivatemanagers, but it can also hamper teamwork, commit-ment, and learning. The way to resolve this dilemma is toapply Theory E incentives in an O way. Employees’ highinvolvement is encouraged to develop their commitmentto change, and variable pay is used to reward that com-mitment. ASDA’s senior executives were compensatedwith stock options that were tied to the company’s value.These helped attract key executives to ASDA. Unlikemost E-strategy companies, however, ASDA had a stock-ownership plan for all employees. In addition, store-levelemployees got variable pay based on both corporate per-formance and their stores’ records. In the end, compen-sation represented a fair exchange of value between thecompany and its individual employees. But Norman

18 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 18

believed that compensation had not played a major rolein motivating change at the company.

Use consultants as expert resources who empoweremployees. Consultants can provide specialized knowl-edge and technical skills that the company doesn’t have,particularly in the early stages of organizational change.Management’s task is figuring out how to use thoseresources without abdicating leadership of the changeeffort. ASDA followed the middle ground between The-ory E and Theory O. It made limited use of four consult-ing firms in the early stages of its transformation. Theconsulting firms always worked alongside managementand supported its leadership of change. However, theirengagement was intentionally cut short by Norman toprevent ASDA and its managers from becoming depen-dent on the consultants. For example, an expert in storeorganization was hired to support the task forceassigned to renew ASDA’s first few experimental stores,but later stores were renewed without his involvement.

By embracing the paradox inherent in simultaneouslyemploying E and O change theories, Norman andLeighton transformed ASDA to the advantage of itsshareholders and employees. The organization wentthrough personnel changes, unit sell-offs, and hierarchi-cal upheaval. Yet these potentially destructive actionsdid not prevent ASDA’s employees from committing tochange and the new corporate culture because Normanand Leighton had won employees’ trust by constantly lis-tening, debating, and being willing to learn. Candidabout their intentions from the outset, they balanced thetension between the two change theories.

By 1999, the company had multiplied shareholdervalue eightfold. The organizational capabilities built byNorman and Leighton also gave ASDA the sustainable


HBR029ch1 8/15/01 1:34 PM Page 19

competitive advantage that Dunlap had been unable tobuild at Scott Paper and that Sigler had been unable tobuild at Champion. While Dunlap was forced to sell ademoralized and ineffective organization to Kimberly-Clark, and while a languishing Champion was sold toUPM-Kymmene, Norman and Leighton in June 1999found a friendly and culturally compatible suitor in Wal-Mart, which was willing to pay a substantial premium forthe organizational capabilities that ASDA had sopainstakingly developed.

In the end, the integration of theories E and O createdmajor change—and major payoffs—for ASDA. Such pay-offs are possible for other organizations that want todevelop a sustained advantage in today’s economy. Butthat advantage can come only from a constant willing-ness and ability to develop organizations for the longterm combined with a constant monitoring of share-holder value—E dancing with O, in an unending minuet.

Change Theories in the New Economy

HISTORICALLY, THE STUDY OF change has beenrestricted to mature, large companies that needed toreverse their competitive declines. But the arguments wehave advanced in this article also apply to entrepre-neurial companies that need to manage rapid growth.Here, too, we believe that the most successful strategyfor change will be one that combines theories E and O.

Just as there are two ways of changing, so there aretwo kinds of entrepreneurs. One group subscribes to anideology akin to Theory E. Their primary goal is to pre-pare for a cash-out, such as an IPO or an acquisition by

20 Beer and Nohria

HBR029ch1 8/15/01 1:34 PM Page 20

an established player. Maximizing market value beforethe cash-out is their sole and abiding purpose. Theseentrepreneurs emphasize shaping the firm’s strategy,structure, and systems to build a quick, strong marketpresence. Mercurial leaders who drive the companyusing a strong top-down style are typically at the helm ofsuch companies. They lure others to join them using high-powered incentives such as stock options. The goal is toget rich quick.

Other entrepreneurs, however, are driven by an ide-ology more akin to Theory O—the building of an institu-tion. Accumulating wealth is important, but it is secondaryto creating a company that is based on a deeply heldset of values and that has a strong culture. Theseentrepreneurs are likely to subscribe to an egalitarianstyle that invites everyone’s participation. They look toattract others who share their passion about the cause—though they certainly provide generous stock options aswell. The goal in this case is to make a difference, notjust to make money.

Many people fault entrepreneurs who are driven bya Theory E view of the world. But we can think of otherentrepreneurs who have destroyed businesses becausethey were overly wrapped up in the Theory O pursuit ofa higher ideal and didn’t pay attention to the pragmaticsof the market. Steve Jobs’s venture, Next, comes to mind.Both types of entrepreneurs have to find some way oftapping the qualities of theories E and O, just as largecompanies do.

Originally published in May–June 2000Reprint R00301


HBR029ch1 8/15/01 1:34 PM Page 21

Turning Goals into ResultsThe Power of Catalytic Mechanisms

Executive Summary

MOST EXECUTIVES HAVE a big, hairy, audacious goal.They write vision statements, formalize procedures, anddevelop complicated incentive programs—all in pursuit ofthat goal. In other words, with the best of intentions, theyinstall layers of stultifying bureaucracy. But it doesn’t haveto be that way.

In this article, Jim Collins introduces the catalytic mech-anism, a simple yet powerful managerial tool that helpstranslate lofty aspirations into concrete reality. Catalyticmechanisms are the crucial link between objectives andperformance; they are a galvanizing, nonbureaucraticmeans to turn one into the other.

What’s the difference between catalytic mechanismsand most traditional managerial controls? Catalyticmechanisms share five characteristics. First, they producedesired results in unpredictable ways. Second, they

23

HBR029ch2 8/15/01 1:34 PM Page 23

distribute power for the benefit of the overall system,often to the discomfort of those who traditionally holdpower. Third, catalytic mechanisms have teeth. Fourth,they eject “viruses”—those people who don’t share thecompany’s core values. Finally, they produce an ongo-ing effect.

Catalytic mechanisms are just as effective for reach-ing individual goals as they are for corporate ones. Toillustrate how catalytic mechanisms work, the authordraws on examples of individuals and organizations thathave relied on such mechanisms to achieve their goals.The same catalytic mechanism that works in one organi-zation, however, will not necessarily work in another.Catalytic mechanisms must be tailored to specific goalsand situations. To help readers get started, the authoroffers some general principles that support the process ofbuilding catalytic mechanisms effectively.

M a big, hairy, audaciousgoal. One dreams of making his brand more popularthan Coke; another aspires to create the most lucrativeWeb site in cyberspace; yet another longs to see her orga-nization act with the guts necessary to depose its archrival. So, too, most executives ardently hope that theiroutsized goals will become a reality. To that end, theywrite vision statements, deliver speeches, and launchchange initiatives. They devise complicated incentiveprograms, formalize rules and checklists, and pen poli-cies and procedures. In other words, with the best inten-tions, they create layer upon layer of stultifying bureau-cracy. Is it any surprise that their wildly ambitiousdreams are seldom realized?

24 Collins

HBR029ch2 8/15/01 1:34 PM Page 24

But companies don’t have to act that way. Over thepast six years, I have observed and studied a simple yetextremely powerful managerial tool that helps organiza-tions turn goals into results. I have recently codified it; I

call it the catalytic mecha-nism. Catalytic mecha-nisms are the crucial linkbetween objectives andperformance; they are agalvanizing, nonbureau-cratic means to turn one

into the other. Put another way, catalytic mechanismsare to visions what the central elements of the U.S. Con-stitution are to the Declaration of Independence—devices that translate lofty aspirations into concrete real-ity. They make big, hairy, audacious goals reachable.

My research indicates that few companies—perhapsonly 5% or 10%—currently employ catalytic mechanisms,and some of them aren’t even aware that they do. I havealso found that catalytic mechanisms are relatively easyto create and implement. Given their effectiveness, theyare perhaps the most underutilized—and most promis-ing—devices that executives can use to achieve their big,hairy, audacious goals, or BHAGs. (For more on BHAGs,see “Anatomy of a BHAG” at the end of this article.)

Consider Granite Rock, a 99-year-old company inWatsonville, California, that sells crushed gravel, con-crete, sand, and asphalt. Twelve years ago, when brothersBruce and Steve Woolpert became copresidents, theygave their company a new BHAG. Granite Rock wouldprovide total customer satisfaction and achieve a reputa-tion for service that met or exceeded that of Nordstrom,the upscale department store that is world famous fordelighting its customers. Not exactly a timid goal for a

Catalytic mechanisms arethe most promising devices executives can use to achieve their big, hairy, audacious goals.

Turning Goals into Results 25

HBR029ch2 8/15/01 1:34 PM Page 25

stodgy, family-owned company whose employees aremostly tough, sweaty people operating rock quarries andwhose customers—mainly tough, sweaty constructionworkers and contractors—are not easily dazzled.

Now stop and think for a minute: What would it taketo actually reach such an ambitious goal? Most peopleautomatically think of galvanizing leadership. But thatwasn’t an option for Granite Rock, as the Woolperts are aquiet, thoughtful, and bookish clan. Nor did the answerlie in hosting hoopla events or launching grand customerservice initiatives. The brothers had seen such efforts atother companies and believed they had little lastingeffect.

They chose instead to implement a radical new policycalled “short pay.” The bottom of every Granite Rockinvoice reads, “If you are not satisfied for any reason,don’t pay us for it. Simply scratch out the line item, writea brief note about the problem, and return a copy of thisinvoice along with your check for the balance.”

Let me be clear about short pay. It is not a refund pol-icy. Customers do not need to return the product. Theydo not need to call and complain. They have completediscretionary power to decide whether and how much topay based on their satisfaction level.

To put the radical nature of short pay in perspective,imagine paying for airline tickets after the flight and hav-ing the power to short pay depending on your travelexperience—not just in the air, but during ticketing anddeplaning as well. Or suppose universities issued tuitioninvoices at the end of the semester, along with the state-ment, “If you are not satisfied with the dedication of theprofessor in any course, simply scratch out that courseand send us a tuition check for the balance.” Or supposeyour cell phone bill came with a statement that said, “Ifyou are not satisfied with the quality of connection of

26 Collins

HBR029ch2 8/15/01 1:34 PM Page 26

any calls, simply identify and deduct those from the totaland send a check for the balance.”

In the years since it was instituted, short pay has hada profound and positive impact on Granite Rock. Itserves as a warning system, providing hard-to-ignorefeedback about the quality of service and products. Itimpels managers to relentlessly track down the rootcauses of problems in order to prevent repeated shortpayments. It signals to employees and customers alikethat Granite Rock is dead serious about customer satis-faction in a way that goes far beyond slogans. Finally, itkeeps Granite Rock from basking in the glory of itsremarkable success.

And it has had success, as has been widely reported.The little company—it has only 610 employees—hasconsistently gained market share in a commodity busi-ness dominated by behemoths, all the while charging a6% price premium. It won the prestigious MalcolmBaldrige National Quality Award in 1992. And its finan-cial performance has significantly improved—fromrazor-thin margins to profit ratios that rival companieslike Hewlett-Packard, which has a pretax return ofroughly 10%. No doubt, short pay was a critical device forturning the Woolpert brothers’ BHAG into a reality.

Five Parts of a Whole

Obviously, not every company should institute short pay.Rather, companies should have catalytic mechanisms aspowerful as short pay. What, then, is the difference betweena catalytic mechanism and most traditional managerialdevices, such as a company’s hiring and compensation poli-cies? Catalytic mechanisms share five distinct characteris-tics. (See the chart “Catalytic Mechanisms: Breaking fromTradition.”) Let’s look at them in turn.


HBR029ch2 8/15/01 1:34 PM Page 27

Catalytic Mechanisms: Breaking from Tradition

Catalytic mechanisms share five distinct characteristics that distinguish them from traditional controls.

A traditional managerial device,control, or mechanism . . .

reduces variation as it enlarges the orga-nization’s bureaucracy.

concentrates power in the hands ofauthorities who can force people to obeytheir commands.

is understood by employees and execu-tives alike as merely an intention.

attempts to stimulate the right behaviorsfrom the wrong people.

has the short-lived impact of a singleevent or a fad.

A catalytic mechanism . . .

produces desired results in unpredictableways.

distributes power for the benefit of theoverall system, often to the great dis-comfort of those who traditionally holdpower.

has a sharp set of teeth.

attracts the right people and ejectsviruses.

produces an ongoing effect.

Examples of catalytic mechanisms

The red flag made a ferociously opinionated CEOlisten to the challenge of an M.B.A. student—improving the knowledge of the whole class,despite the unexpected nature of the exchange.

A new government rule allowed a low-level man-ager to expunge an immensely wasteful regulationthat required nearly new uniforms to be burned.

Short pay at Granite Rock allows customers to payonly for the products that satisfy them.

At W.L. Gore & Associates, employees can, ineffect, fire their bosses, ensuring nonhierarchicalleadership.

Kimberly-Clark knowingly put itself into head-to-head competition with Procter & Gamble to impelbetter performance in the consumer goods market-place. Such a strategy is still working 30 years later.

HBR029ch2 8

/15/01 1

:34 PM P

age 28

Characteristic 1: A catalytic mechanism producesdesired results in unpredictable ways. When execu-tives identify a bold organizational goal, the first thingthey usually do is design a plethora of systems, controls,procedures, and practices that seem likely to make ithappen. That process is called alignment, and it’s wildlypopular in the world of management, among businessacademics and executives alike. After all, alignmentmakes sense. If you want to make your brand more pop-ular than Coke, you had better measure the effectivenessof advertising and reward successful marketing man-agers with big bonuses. But the problem, as I’ve said, isthat the controls that undergird alignment also createbureaucracy, and it should be news to no one thatbureaucracy does not breed extraordinary results.

Don’t get me wrong. Bureaucracy may deliver results,but they will be mediocre because bureaucracy leads topredictability and conformity. History shows us thatorganizations achieve greatness when people are allowedto do unexpected things—to show initiative and creativ-ity, to step outside the scripted path. That is whendelightful, interesting, and amazing results occur.

Take 3M. For decades, its executives have dreamed ofhaving a constant flow of terrific new products. Toachieve that end, in 1956, the company instituted a cat-alytic mechanism that is by now well known: scientistsare urged to spend 15% of their time experimenting andinventing in the area of their own choice. How veryunbureaucratic! No one is told what products to workon, just how much to work. And that loosening of con-trols has led to a stream of profitable innovations, fromthe famous Post-it Notes to less well-known examplessuch as reflective license plates and machines thatreplace the functions of the human heart during surgery.


HBR029ch2 8/15/01 1:34 PM Page 29

3M’s sales and earnings have increased more than 40-fold since instituting the 15% rule. The mechanism hashelped generate cumulative stock returns 36% in excessof the market and has earned the company a frequentranking in the top ten of Fortune’s most-admired list.

In a happy coincidence, the variation sparked by cat-alytic mechanisms forces learning to occur. Suppose youset out to climb the 3,000-foot sheer rock face of El Capi-tan in Yosemite Valley. Once you pass pitch 15, you can-not possibly retreat from your particular route: you are,by dint of nature, 100% committed. Although you can’tpredict how you will overcome the remaining pitches—you have to improvise as you go—you can predict thatyou will invent a way to the top. Why? Because the real-ity of having no easy retreat forces you to reach the sum-mit. Catalytic mechanisms have the same effect. GraniteRock’s short pay commits the company to achievingcomplete customer satisfaction. Every time a customerexercises short pay, Granite Rock learns or invents a wayto run its operations more effectively. Ultimately, suchnew knowledge leads to better results, making the cat-alytic mechanism part of a virtuous circle of variation,learning, improvement, and enhanced results.

My “red flag” device also illustrates that circle. When Ifirst began teaching Stanford M.B.A. students by the casemethod in 1988, I noticed that a small number of themtended to dominate the discussion. I also noticed thatthere was no correlation between the degree of vocalaggressiveness and how much these students improvedthe class’s overall learning experience. Some vocal stu-dents had much to contribute; others just liked to hearthemselves talk. Worse, I noticed when chatting withstudents after class that some of the quieter individualshad significant contributions but were selective or shyabout sharing them. Furthermore, seeing 15 to 20 hands

30 Collins

HBR029ch2 8/15/01 1:34 PM Page 30

raised at a time, I had no way of knowing which one rep-resented a truly significant insight, and I sensed that Iwas frequently missing some students’ one best contri-bution for the entire quarter.

I solved that problem by giving each student an 8.5inch by 11 inch bright red sheet of paper at the beginningof every quarter. It had the following instructions: “This isyour red flag for the quarter. If you raise your hand withyour red flag, the classroom will stop for you. There are norestrictions on when and how to use your red flag; thedecision rests entirely in your hands. You can use it tovoice an observation, share a personal experience, presentan analysis, disagree with the professor, challenge a CEOguest, respond to a fellow student, ask a question, make asuggestion, or whatever. There will be no penalty whatso-ever for any use of a red flag. Your red flag can be usedonly once during the quarter. Your red flag is nontransfer-able; you cannot give or sell it to another student.”

I had no idea precisely what would happen each dayin class. And yet, the red flag device quickly created abetter learning experience for everyone. In one case, itallowed a very thoughtful and quiet student from Indiato challenge Anita Roddick on the Body Shop’s manufac-turing practices in the Third World. Roddick, a charis-matic CEO with ferociously held views, usually domi-nates any discussion. The red flag forced her to listen toa critic. The spirited interchange between these two pas-sionate and well-informed people produced more learn-ing than anything I could have scripted. Without the redflag, we would have just had another session of “I’m CEOand let me tell you how it is.”

In another situation, a student used her red flag tostate, “Professor Collins, I think you are doing a particu-larly ineffective job of running class today. You areleading too much with your questions and stifling our


HBR029ch2 8/15/01 1:34 PM Page 31

independent thinking. Let us think for ourselves.” Thatwas a tough moment for me. My BHAG as a professorwas to create the most popular class at the businessschool while imposing the highest workload and thestiffest daily standards. The red flag system confrontedme with the fact that my own questioning style stood inthe way of my dream—but it also pointed the way toimprovement, again, to everyone’s benefit.

Interestingly, no other professors on campus adoptedthe red flag. One of them told me, “I can’t imagine doingthat. I mean, you never know what might happen. I couldnever give up that much control in my classroom.” Whathe and others missed was a great paradox: by giving upcontrol and decreasing predictability, you increase theprobability of attaining extraordinary results.

Characteristic 2: A catalytic mechanism distributespower for the benefit of the overall system, often tothe great discomfort of those who traditionallyhold power. With enough power, executives can alwaysget people to jump through hoops. If it is customer ser-vice they are after, for instance, they can threaten dis-missal to coerce salespeople to smile and act friendly. Ifthey seek higher profits per store, they can pay employ-ees according to flow-through. And if increased marketshare is the dream, they can promote only those man-agers who make it happen.

But consider how catalytic mechanisms work. Shortpay distributes power to the customer, to the great dis-comfort of Granite Rock’s executives, but toward thegreater goal of continuous improvement for the benefitof customers and company alike. The red flag distributespower to the students, to the great discomfort of theteacher, but to the ultimate improvement of learning in

32 Collins

HBR029ch2 8/15/01 1:34 PM Page 32

general. The founders of the United States understoodthis point when they wrote the Constitution. After all,the Constitution is the set of catalytic mechanisms thatreinforce and support the national vision. Voting, thesystem of checks and balances, the two-thirds vote toamend, the impeachment process—these disperse poweraway from one central source, to the great discomfort ofthose who seek power, but to the benefit of the overallnation.

Catalytic mechanisms force the right things to hap-pen even though those in power often have a vestedinterest in the right things not happening. Or they have avested interest in inertia—letting pointless, expensivepractices stay in place. That’s what happened for years,perhaps decades, at U.S. Marine recruit depots. Allrecruits are issued a uniform on their first day. Twoweeks later, they need another—the pounds melt awaywhen you run 12 miles every dawn. The military’s rulesrequired those two-week-old uniforms to be destroyed.Not washed and reissued, but destroyed.

In the early 1990s, Phil Archuleta, a materials man-ager at a recruiting depot in San Diego, suggested that

they reuse the uniforms.His boss’s response: “No.It’s against regulations.Forget about it.” So in afabulous act of insubordi-nation, Archuleta washedthe uniforms, hid them in

boxes, and bided his time until he finally got a supervisorwilling to challenge the regulation.

In an effort to empower the Phil Archuletas of theworld, the government launched a wide-ranging initia-tive in 1994 to fix its bureaucratic quagmire. A new rule

Catalytic mechanismssubvert the default knee-jerk tendency of bureaucraciesto choose status quoover change.


HBR029ch2 8/15/01 1:34 PM Page 33

regarding waivers was put in place, and it is a catalyticmechanism that exemplifies the beauty and power ofredistributing power. It has two primary components:

• Waiver-of-regulation requests must be acted uponwithin 30 days. After 30 days, if no answer is forth-coming, the party asking for the waiver can assumeapproval and implement the waiver.

• Those officials who have the authority to change regu-lations can approve waiver requests, but only the headof an agency can deny a request.

Think for a minute about the impact of this catalyticmechanism. It subverts the default, knee-jerk tendency ofbureaucracies to choose inaction over action, status quoover change, and idiotic rules over common sense. Super-visors can no longer say no or not respond. They wouldhave to champion a no all the way to the head of theiragency—the equivalent of the head commandant of theentire U.S. Marine Corps—within 30 days. Instead of hav-ing to go out of their way to demonstrate why it is a goodidea, they would have to expend great energy to provethat it is a bad idea. The catalytic mechanism tilts thebalance of power away from inertia and toward change.

Indeed, the primary effect of the new waiver rule—aswith all catalytic mechanisms—is to give people the free-dom to do the right thing. The waiver that allowedArchuleta to change the regulation on uniforms createda savings of half a million dollars in two years. Similarexamples of people doing the right thing with the waiverrule abound throughout the federal government, fromthe FDA to NASA. Tort claims adjusters in the Depart-ment of Agriculture, for instance, waived regulations toreduce processing time of claims from 51 days to eight

34 Collins

HBR029ch2 8/15/01 1:34 PM Page 34

days—a manpower savings of 84%. When executives vestpeople with power and responsibility and step out of theway, vast reservoirs of energy and competence flow forth.Again we have a paradox: the more executives dispersepower and responsibility, the more likely the organiza-tion is to reach its big, hairy, audacious goal.

Characteristic 3: A catalytic mechanism has teeth.Lots of companies dream of total customer satisfaction;few have a device for making it happen that has the teethof short pay. Plenty of organizations state the lofty inten-tion to empower people; few translate that into resultswith a mechanism that has the teeth of the red flag.Many companies state that they intend to “become num-ber one or number two in every competitive arena”; fewhave added an effective means of enforcement by saying,“and if the business is not number one or number two, oron a clear trajectory to get there, we will exit within threemonths.”

The fact is, executives spend hours drafting, redraft-ing, and redrafting yet again statements of core values,missions, and visions. This is often a very useful process,but a statement by itself will not accomplish anything.By contrast, a catalytic mechanism puts a process inplace that all but guarantees that the vision will be ful-filled. A catalytic mechanism has a sharp set of teeth.

Consider the case of Nucor Corporation, the mostsuccessful U.S. steel company of the last three decades. Ithas a unique vision for a Rust Belt company: to be anorganization whose workers and management share thecommon goal of being the most efficient, high-qualitysteel operation in the world, thereby creating job securityand corporate prosperity in an industry ravaged byforeign competition. Behind that vision lies the belief


HBR029ch2 8/15/01 1:34 PM Page 35

held deeply by Nucor’s senior leaders that decent, hard-working people should be well paid for their efforts and,so long as they are highly productive, that they need notworry about job security.

On the surface, Nucor’s vision may sound warm andfuzzy. Dig deeper, and you’ll see that it actually leaves noroom for unproductive employees. Nucor has created aculture of intense productivity whereby five people dothe work that ten do at other steel companies, and getpaid like eight. The vision came to life through a series ofpowerful catalytic mechanisms with teeth, such as theway frontline workers get paid:

• Base hourly pay is 25% to 33% below the industryaverage.

• People work in teams of 20 to 40; team-productivityrankings are posted daily.

• A bonus of 80% to 200% of base pay, based on teamproductivity, is paid weekly to all teams that meet orexceed productivity goals.

• If you are five minutes late, you lose your bonus forthe day.

• If you are 30 minutes late, you lose your bonus for theweek.

• If a machine breaks down, thereby stopping produc-tion, there is no compensating adjustment in thebonus calculation.

• If a product is returned for poor quality, bonus paydeclines accordingly.

You might be thinking that the Nucor system concen-trates power in the hands of management, which wouldseem to contradict the idea of distributing power for the

36 Collins

HBR029ch2 8/15/01 1:34 PM Page 36

sake of the system. But in fact, the catalytic mechanismactually takes the power out of the hands of individualmanagers and their whims. Nucor has no discretionarybonuses. It’s more like a sports bonus system: if youscore so many points or win a certain number of races,you get a bonus based on a predetermined formula.Period. That formula gives workers more power overtheir own destiny than bonus programs that give largediscretionary power to management. If your team scoresthe points, your team gets the bonus, and no managercan take it away, citing, “We’re just not having a verygood year” or “I don’t like your attitude.”

Nucor’s catalytic mechanisms for managers, inciden-tally, have even sharper teeth. Its executive compensa-tion system works very much like its worker compensa-tion system, except that the “team” is the entire plant

(for plant managers) or theentire company (for corpo-rate officers). And, unlikemost companies, whentimes are bad, Nucor’sexecutives assume greaterpain than frontline work-

ers: workers’ pay drops about 25%, plant managers’ paydrops about 40%, and corporate officers’ pay drops about60%. In the 1982 recession, CEO Ken Iverson’s paydropped 75%.

Characteristic 4: A catalytic mechanism ejectsviruses. A lot of traditional controls are designed to getemployees to act the “right” way and do the “right”things, even if they are not so inclined. Catalytic mecha-nisms, by contrast, help organizations to get the rightpeople in the first place, keep them, and eject those whodo not share the company’s core values.

The old adage “Peopleare your most important asset” is wrong. The right people are your mostimportant asset.


HBR029ch2 8/15/01 1:34 PM Page 37

Great organizations have figured something out. Theold adage “People are your most important asset” iswrong; the right people are your most important asset.The right people are those who would exhibit the desiredbehaviors anyway, as a natural extension of their charac-ter and attitude, regardless of any control and incentivesystem. The challenge is not to train all people to shareyour core values. The real challenge is to find people whoalready share your core values and to create catalyticmechanisms that so strongly reinforce those values thatthe people who don’t share them either never get hiredor, if they do, they self-eject.

Let’s return to the Nucor example. Nucor doesn’t tryto make lazy people productive. Its catalytic mechanismscreate a high-performance environment in which thosewith an innate work ethic thrive and free riders get outin a hurry. Management usually doesn’t fire unproduc-tive workers; workers do. In one case, team memberschased a lazy coworker out of the plant. And onereporter writing a story on Nucor described showing upfor a shift on time but thinking he was late because allthe workers had been there for 30 minutes arrangingtheir tools and getting ready to fire off the starting lineprecisely at 7:00 a.m.

Interestingly, Nucor sets up its mills not in traditionalsteel towns, but primarily in rural, agricultural areas. Thethinking is simple: you can’t teach the work ethic—either a person has it or he doesn’t. But you can teachsteel making. That’s why Nucor hires farmers and trainsthem. The company’s catalytic mechanisms wouldn’thave it any other way.

Another example of a catalytic mechanism ejectingviruses comes from W.L. Gore & Associates, a fabriccompany worth nearly $2 billion. Bill Gore founded the

38 Collins

HBR029ch2 8/15/01 1:34 PM Page 38

company in 1958 with the vision of creating a culture ofnatural leadership. Leadership, in Gore’s view, could notbe assigned or bestowed by hierarchical position. You area leader if and only if people choose to follow you. Gore’stheory sprang not just from his personal values but alsofrom his business sense: he thought that the most cre-ative and productive work came when people freelymade commitments to one another, not when bossestold them what to do.

To turn his vision into reality, Gore invented a cat-alytic mechanism that attracted the right people like amagnet and scared away the others. At W.L. Gore &Associates, employees have the authority to fire theirbosses. Now, they can’t fire the person from the companybut, if they feel their boss isn’t leading them effectively,they can simply bypass him or her and follow a differentleader.

Who would want to work at such a company? Exactlythe people who belong there—people who know they canlead without the crutch of a formal position or title andwho believe in the philosophy of nonhierarchical leader-ship. Who would avoid it like the plague? Anyone whogets giddy pulling the levers of position and power justfor the pulling’s sake. And if you’re a hierarchical leaderwho happens to make it through the company’s door butcan’t quickly shake the notion that “the boss has to bethe boss,” it won’t take you long to find the exit.

Characteristic 5: A catalytic mechanism producesan ongoing effect. Catalytic mechanisms differ funda-mentally from catalytic events. A rousing speech to thetroops, an electrifying off-site meeting, a euphoria-producing new buzzword, a new initiative or strategicimperative, an impending crisis—all of these are


HBR029ch2 8/15/01 1:34 PM Page 39

catalytic events, and some are useful. But they do notproduce the persistent, ongoing effect of catalyticmechanisms. In fact, a good catalytic mechanism, aslong as it evolves, can last for decades, as the 15% ruleat 3M and the impeachment mechanism in the Consti-tution illustrate.

The lack of catalytic mechanisms is one reason manyorganizations rally in a crisis but languish once the crisishas passed. Leaders who feign a crisis—those who createa burning platform without simultaneously building cat-alytic mechanisms—do more long-term harm than goodby creating a syndrome of crisis addiction. Executiveswho rely only on catalytic events are left wondering whythe momentum stalls after the first phase of euphoria,excitement, or fear has passed. To produce lastingresults, they must shift from orchestrating a series ofevents to building catalytic mechanisms.

Take, for example, the decades of ineffectual attemptsto reform public education in the United States. Part ofthe failure lies in the approach to reform; too often it isbased on onetime events and fashionable buzzwordsrather than on catalytic mechanisms that produce sus-tained effects. As Roger Briggs, a high school teacher inBoulder, Colorado, wrote in an essay on school reform:“Every year we get a new program or fad. And they neverreally work. And we teachers eventually just learn to ig-nore them, smile, and go about our business of teaching.”

Now take a look at what happened when the state ofTexas started using a catalytic mechanism in 1995: com-parison-band ranking of schools, which is directly tied toresource allocation and, in some cases, school closures.The ongoing effect of this device forced the momentum ofreform forward. Why? Well, if you rank fifth out of 40schools but you just sit still, you’ll drop in the ratings. Sitstill long enough, and you’ll eventually rank 35th rather

40 Collins

HBR029ch2 8/15/01 1:34 PM Page 40

than fifth, and you may face closure. Because every schoolis ranked on the same criteria, the bar for performancekeeps rising. Within four years of installing the mecha-nism, student achievement in Texas improved across theboard. The percentage of students who passed the Texasmath skill exam, for example, rose from roughly half to80%, and the share of black and Hispanic students whopassed doubled to 64% and 72%, respectively.

And consider the ongoing impact of a good catalyticmechanism in a more corporate setting. Darwin Smith,former CEO of Kimberly-Clark, created in 1971 theBHAG to transform Kimberly-Clark from a mediocre for-est- and paper-products company into a world-class con-sumer goods company. At the time, Wall Street analystsscoffed at the idea, as did most of Kimberly-Clark’s com-petitors. Smith was undeterred. He created one catalyticevent and one equally important catalytic mechanism.For the first, he sold a big chunk of the company’s tradi-tional paper-production mills, thus leaving no easyescape route from the dream. For the second, he com-mitted the company to head-to-head competition withthe best consumer-products company in the world: Proc-ter & Gamble. With its entry into disposable diapers,Kimberly-Clark would henceforth be a direct rival ofP&G. Kimberly-Clark would either become excellent atconsumer products or get crushed. The beauty of thiscatalytic mechanism is that, unlike the “change or die”ranting all too common among modern executives, itsongoing effect is as powerful today as when it was firstput in place nearly 30 years ago.

Getting Started

This is not intended to be a how-to article; my mainobjective has been to introduce the concept of catalytic


HBR029ch2 8/15/01 1:34 PM Page 41

mechanisms and demonstrate how they have helpedsome companies—and individuals—turn their BHAGsinto reality. (For more on the personal use of catalyticmechanisms, see “Not for Companies Only” at the end ofthis article.) Nonetheless, my research suggests thatthere are a few general principles that support the pro-cess of building catalytic mechanisms effectively.

’ ,

When pursuing BHAGs, our natural inclination is toadd—new initiatives, new systems, new strategies, newpriorities, and now, new catalytic mechanisms. But indoing so, we overwhelm ourselves. Isn’t it frightening thatthe new version of the Palm Pilot has space for 1,500 itemson its to-do list? Sadly, few of us have a “stop doing” list.We should, because to take something away—to unplugit—can be as catalytic as adding something new.

Take the case of a circuit division at Hewlett-Packard.It had tried countless programs and initiatives to reachits BHAG of becoming “a place where people would walk

on the balls of their feet,feel exhilarated about theirwork, and search for imagi-native ways to improve andinnovate everything wedo.” The events producedshort-term results—a

moment of sparkle and excitement—but within a monthor two, the division always drifted back into its sleepy,humdrum mode.

Then its executives considered the question, “Whatpolicies should we remove?” For most of its history, thedivision had comfortably lived off a captive internal mar-

A catalytic mechanism should be an idiosyncratic adaptation, if not a wholesale creation, for a unique situation.

42 Collins

HBR029ch2 8/15/01 1:34 PM Page 42

ket. What if HP’s divisions were allowed to buy theircomponents from outside competitors? Never againwould the circuit division have fat internal orders justhanded to it. Never again could it just sit still. Twomonths, four months, a year, five years, and ten yearsdown the road—fierce competitors would still be there,constantly upping the ante. The prospect was both terri-fying and exhilarating. Managers decided to unplug the“buy internal” requirement and open the doors to free-market competition.

Within weeks, the circuit division was well on its wayto realizing its BHAG. You could sense a completely dif-ferent environment the moment you walked in the door.The place hummed with activity, and its performanceshowed it.

, ’

Creating mechanisms is exactly that: a creative act. Youcan, of course, get good ideas by looking at what otherorganizations do, but the best catalytic mechanisms areidiosyncratic adaptations, if not wholesale creations, fora unique situation.

Because catalytic mechanisms require fresh ideas, itmakes sense to invite all members of an organization toparticipate in their creation. Everyone. Certainly, somemechanisms require input from senior executives, likeshort pay at Granite Rock. Yet many of the best catalyticmechanisms were not created by top management. Theidea for the federal government’s waiver rule, for exam-ple, originated with two staff members—Lance Cope andJeff Goldstein. They were working in the national rein-vention labs, and neither had direct authority over anyfederal agency.


HBR029ch2 8/15/01 1:34 PM Page 43

Allow me also to use a personal example. Part of myprofessional vision is to contribute through teaching andto harness my curiosity and passion for learning in waysthat make a positive impact on the world. From that goalflows the imperative that I allocate time primarily toresearch, writing, and teaching and limit consultingwork only to those situations in which I can contributeas a teacher.

To reinforce that imperative, I have created two cat-alytic mechanisms: the “come to Boulder rule” and the“four-day rule.” The first rule states that I will not engagein a direct advisory relationship with any organizationunless the chief executive agrees to travel to my Boulderresearch laboratory. Executives spend huge sums ofmoney on consultants, but money doesn’t equal commit-ment—if you have a big enough budget, invoices justdon’t hurt. Yet all chief executives, no matter how largetheir budgets, have only 24 hours in a day. If a CEO fliesall the way to Boulder, he or she has demonstrated com-mitment to serious discussions and hard work, and thelikelihood that I will make a significant impact as ateacher increases exponentially. Most important, thosenot committed to real (and perhaps uncomfortable)change eject right up front.

The second mechanism—my four-day rule—statesthat any given organization has an upper limit of fourdays of my advisory time in a year. The most lastingimpact comes by teaching people how to fish, not by fish-ing for them. Organizations that want an adviser to fishfor them self-eject through this catalytic mechanism.Admittedly, these are highly unusual devices, and theywould be disastrous for most consulting firms thatdepend on continuous growth to feed their machine. Yet

44 Collins

HBR029ch2 8/15/01 1:34 PM Page 44

they are perfectly designed for a strategy aimed at explic-itly not building a large consulting business. They areunique to me, as all catalytic mechanisms should be totheir creators.

,

The examples in this article may lead you to believe thatmost catalytic mechanisms use money. But, in fact, whenmy research colleague Lane Hornung cataloged mydatabase of catalytic mechanisms, he found that onlyhalf do. That might surprise some people—in particularthose who ascribe to the old saw that money is the bestmotivator. I’m not going to claim that money doesn’timpel people toward desired results; money can addteeth to any catalytic mechanism. But to rely entirely onmoney reflects a shallow understanding of humanbehavior.

The U.S. Marine Corps illustrates my point precisely.The Corps builds extraordinary commitment through aset of catalytic mechanisms that create intense psycho-logical bonds among its members. By isolating recruits atboot camps and creating an environment where recruitssurvive only by relying upon one another, the Corps trig-gers the deep human drive, hardwired into most of us, tosupport and protect those we consider family. Most peo-ple will not risk their lives for a year-end bonus, but theywill go to great lengths to earn the respect and protectthe well-being of their comrades.

William Manchester, who returned to his unit on Oki-nawa after receiving a wound that earned him a PurpleHeart, eloquently describes the psychology of commit-ment in his book Goodbye Darkness:


HBR029ch2 8/15/01 1:34 PM Page 45

And then, in one of those great thundering jolts in which aman’s real motives are revealed to him in an electrifyingvision, I understand, at last, why I jumped hospital thatSunday thirty-five years ago, and, in violation of orders,returned to the front and almost certain death. It was anact of love. Those men on the line were my family, myhome. . . . They had never let me down, and I couldn’tdo it to them. I had to be with them rather than to letthem die and me live with the knowledge that I mighthave saved them. Men, I now knew, do not fight for flag orcountry, for the Marine Corps or glory or any otherabstraction. They fight for one another.1

Yes, catalytic mechanisms sometimes use money toadd bite, but the best ones also tap deeper wells ofhuman motivation. Even at Nucor, the effectiveness of itscatalytic mechanisms lies as much in the peer pressureand the desire to not let teammates down as in the num-ber of dollars in the weekly bonus envelope. The bestpeople never work solely for money. And catalytic mech-anisms should reflect that fact.

New catalytic mechanisms sometimes produce unin-tended negative consequences and need correction. Forinstance, the first version of the red flag failed becausecertain students continued to dominate class discussion,thinking that every comment of theirs was worth a redflag. So I added the stipulation: “Your red flag can be usedonly once during the quarter. Your red flag is nontransfer-able; you cannot give or sell it to another student.”

All catalytic mechanisms, in fact, even if they workperfectly at first, should evolve. 3M’s 15% rule is a case inpoint. In 1956, executives urged 3M scientists to use 3M

46 Collins

HBR029ch2 8/15/01 1:34 PM Page 46

labs during their lunch break to work on anything theywanted. In the 1960s, that catalytic mechanism becameformalized as the “15% rule,” whereby scientists coulduse any 15% of their time. In the 1980s, the 15% rulebecame widely available to 3Mers other than scientists,to be used for manufacturing and marketing innova-tions, for example. In the 1990s, 3M’s executives worriedthat fewer people were using the mechanism than in pre-vious decades. It put together a task force to reinvent the15% rule, bolstering it with special recognition rewardsfor those who used their “bootleg time”—as it has cometo be called—to create profitable innovations.

The 15% rule has been a catalytic mechanism at 3Mfor more than 40 years, but it has continually evolved inorder to remain relevant and effective. That’s the rightapproach; no catalytic mechanism should be viewed assacred. In a great company, only the core values and pur-pose are sacred; everything else, including a catalyticmechanism, should be open for change.

One catalytic mechanism is good; several that reinforceone another as a set is even better. That’s not to say acompany needs hundreds of catalytic mechanisms—ahandful will do. Consider Granite Rock again. It certainlydoesn’t rely just on short pay. It also has a catalytic mech-anism that requires an employee and manager to create afocused development plan for the employee during theperformance evaluation process. Indeed, every employeeand manager must together complete a form that reads:“Learn_____ so that I can contribute_____.” Two sets ofteeth make this form effective. First, employees and theirmanagers must both sign off on the final development


HBR029ch2 8/15/01 1:34 PM Page 47

plan, which forces a continual dialogue until they reachagreement. Second, compensation ties directly to learn-ing and improvement, not just job performance: peoplewho do not go out of their way to improve their skills re-ceive lower than midpoint pay. Only those who do a goodjob and improve their skills and make a contribution toimproving the overall Granite Rock system receive higherthan midpoint pay. So people who merely do a good jobself-eject out of Granite Rock. This catalytic mechanismhas produced delightful surprises: one previously illiter-ate employee used it to get the company to send him to areading program. When Granite Rock won the BaldrigeAward, he read an acceptance speech.

Granite Rock also uses catalytic mechanisms to guidehiring, encourage risk taking, and stimulate new capabil-ities. The point here is not so much in the details as it isin the big picture: Granite Rock does not rely solely onshort pay to pursue its BHAG of attaining a reputationfor customer satisfaction that exceeds Nordstrom’s. Ithas about a dozen catalytic mechanisms that supportand reinforce one another.

That said, however, it would be a mistake to take thisarticle and launch a grand catalytic mechanism initia-tive. Developing a set of catalytic mechanisms should bean organic process, an ongoing discipline, a habit ofmind and action. The dozen or so catalytic mechanismsat Granite Rock came into being over a ten-year period.You certainly don’t want to use the idea to createanother layer of bureaucracy. Catalytic mechanismsshould be catalysts, not inhibitors.

Castles in the Air

I recently worked with a large retail chain to define itsBHAG for the twenty-first century. The company is

48 Collins

HBR029ch2 8/15/01 1:34 PM Page 48

doing well, but it wants its performance to be outra-geously great. And so its executives came up with awildly ambitious goal: to make its brand more popularthan Coke.

That company’s challenge now is to invent the cat-alytic mechanisms that will make the dream a reality.I’ve advised its executives against investing heavily inhoopla events to fire up thousands of frontline employ-ees about the new BHAG. Instead, they should createand implement a set of catalytic mechanisms—specific,concrete, and powerful devices to lend discipline to theirvision. After all, catalytic mechanisms alone will not cre-ate greatness; they need a dream to guide them. But ifyou can blend huge, intangible aspirations with simple,tangible catalytic mechanisms, then you’ll have themagic combination from which sustained excellencegrows.

At the conclusion of Walden, Henry David Thoreauwrote: “If you have built castles in the air, your workneed not be lost; that is where they should be. Now putthe foundations under them.” BHAGs are a company’swildest dreams. Catalytic mechanisms are their founda-tions. Build them both.

Anatomy of a BHAG

IN OUR RESEARCH FOR Built to Last, Jerry Porras and Idiscovered that most enduring great companies set andpursue BHAGs (pronounced BEE-hags and shorthand forbig, hairy, audacious goals). There are three key charac-teristics of a good BHAG:

1. It has a long time frame—ten to 30 years or more. Thewhole point of a BHAG is to stimulate your organization


HBR029ch2 8/15/01 1:34 PM Page 49

to make changes that dramatically improve its fundamen-tal capabilities over the long run. Citicorp’s first BHAG,set in 1915—to become the most powerful, the most ser-viceable, the most far-reaching world financial institutionever—took more than five decades to achieve. Its newBHAG, set in the early 1990s—to attain 1 billion cus-tomers worldwide—will require at least two decades toachieve. (Today it has less than 100 million.) BHAGswith short time frames can lead executives to sacrificelong-term results for the sake of achieving a short-termgoal.

2. It is clear, compelling, and easy to grasp. The goalin a good BHAG is obvious, no matter how youphrase it. For example, Philip Morris’s BHAG, set in the1950s—to knock off R.J. Reynolds as the number onetobacco company in the world—didn’t leave muchroom for confusion. I call this the “Mount Everest stan-dard.” The goal to climb Mount Everest can be said as“Climb the most famous mountain in the world” or“Climb the biggest mountain in the world” or “Climbthe mountain at 87 degrees east, 28 degrees north” or“Climb the mountain in Nepal that measures 29,028feet” or hundreds of other ways. If you find yourselfspending countless hours tinkering with a statement, youdon’t yet have a BHAG.

3. It connects to the core values and purpose of theorganization. The best BHAGs aren’t random; they fitwith the fundamental core values and reason for beingof the company. For example, Nike’s BHAG in the1960s—to crush Adidas—fit perfectly with Nike’s corepurpose “to experience the emotion of competition, win-ning, and crushing competitors.” Sony’s BHAG in the1950s—to become the company most known for chang-ing the worldwide poor-quality image of Japanese prod-

50 Collins

HBR029ch2 8/15/01 1:34 PM Page 50

ucts—flowed directly from its stated core value of elevat-ing the Japanese culture and national status.

This last criterion connects back to the reason for hav-ing a BHAG in the first place. It is a powerful way tostimulate progress—change, improvement, innovation,and renewal—while simultaneously preserving your corevalues and purpose. It is this remarkable ability to blendcontinuity with change that separates enduring greatcompanies from merely successful ones. The trick, ofcourse, is not just to set a BHAG but to achieve it, andtherein lies the power of catalytic mechanisms.

Not for Companies Only

MY RESEARCH HAS FOCUSED on the impact of cat-alytic mechanisms in organizational settings—on how theycan turn a company’s most ambitious goals into reality.But catalytic mechanisms can also have a powerfulimpact on individuals. Indeed, I have made catalyticmechanisms a fundamental part of how I manage mytime, with my “come to Boulder rule” and “four-day rule.”

I am not alone. Several of my former students at Stan-ford Business School have applied a catalytic mecha-nism to reach their goals. In one case, a studentemerged from his courses on entrepreneurship fired upby the idea of forgoing the traditional path and strikingout on his own. But as time passed and he felt the crush-ing burden of school debt as well as the lure of lucrativejob offers, his personal vision waned. He took a job at alarge, established disc drive manufacturer and promisedhimself, “I’m going to launch out on my own in five yearswhen I’ve paid off all my school debts.”


HBR029ch2 8/15/01 1:34 PM Page 51

In most cases, such dreams fade as the years go by—with the advent of cars, houses, children, and all the rest.My former student, however, implemented an interestingcatalytic mechanism to keep his vision alive. He drafteda resignation letter and dated it five years out. Then hegave copies of the letter to a handful of reliable people,along with the following instructions, “If I don’t leave myjob and launch out on my own by the specified date,then send the letter in for me.” His plan worked. In 1996,I received an e-mail from him that described how hesaved his money and spent his off-hours developing hisentrepreneurial options. Then, right on schedule, he quithis secure job and launched a fund to buy and run hisown company.

In another case, a former student created a personalboard of directors composed of people he admires andwould not want to disappoint, and he made a personalcommitment to follow the board’s guidance—it has powerin his life. In 1996, he wrote me: “I recently used my per-sonal board in deciding whether to leave Morgan Stan-ley and go to work with a friend in his two-year-old busi-ness. ‘Yes’ was the unanimous vote.” So despite the riskof leaving a lucrative and prestigious position, he leaptinto the small company, which has since grown fourfoldto employ more than 80 people.

Consider also the highly effective catalytic mecha-nism that a colleague of mine has been using for the pastthree years to attain her BHAG: to lead a full and activelife as a mother, wife, professional writer, and church vol-unteer, without going crazy. That part about maintainingsanity is important because before her catalytic mecha-nism was in place, my colleague constantly found herselfoverextended and miserable. The main culprit was her

52 Collins

HBR029ch2 8/15/01 1:34 PM Page 52

work as a freelance writer: she accepted too many jobs.“Even if we didn’t need the money, I would still take onevery project that came my way,” she recalls. “Maybebecause my family was so poor when I was growing up,I just found it impossible to leave money on the table.”Not surprisingly, the woman’s children paid the price ofher constant working, as did her husband and close-knitextended family. “Either I was too exhausted to see peo-ple or else I was calling them for a baby-sitting favor,”she says.

One day, my colleague was lamenting her situationto her sister, who came up with an effective catalyticmechanism. Every time the woman took on work beyonda certain level of revenue—a comfortable annual salary,in essence—she would pay her sister a $200-a-daypenalty fee. My colleague, instantly seeing the wonder-ful impact of the plan, immediately agreed.

Since she redistributed power to her sister, my col-league has gained new control over her life. Now shehappily accepts jobs up to a certain level of income, butshe assesses each additional offer with newly criticaleyes. (She has taken on extra work on only two occa-sions; both projects were too lucrative to pass up.)Indeed, the catalytic mechanism has so freed my col-league from overwork that she has taken on a new roleas a volunteer at her children’s school. With its undeni-able bite, my colleague’s catalytic mechanism will havean ongoing effect as long as she honors it. And given itsresults, she plans to do so for a long time.

Would any of these people have changed their liveswithout catalytic mechanisms? Perhaps, but I think it lesslikely. Personal catalytic mechanisms have all the benefitsof organizational mechanisms: they put bite into good


HBR029ch2 8/15/01 1:34 PM Page 53

intentions, dramatically increasing the odds of actuallybeing true to your personal vision instead of letting yourdreams remain unrealized.

Notes

1. William Manchester, Goodbye Darkness (Boston: Little,Brown Company, 1979).

Originally published in July–August 1999Reprint 99401

54 Collins

HBR029ch2 8/15/01 1:34 PM Page 54

Changing the Way We Change , ,

Executive Summary

MORE AND MORE COMPANIES struggle with growingcompetition by introducing improvements into everyaspect of performance. But the treadmill keeps movingfaster, the companies keep working harder, and resultsimprove slowly or not at all.

The problem here is not the improvement programs.The problem is that the whole burden of change typicallyrests on so few people. Companies achieve real agilityonly when every function and process—when every per-son—is able and eager to rise to every challenge. Thistype and degree of fundamental change, commonlycalled revitalization or transformation, is what many com-panies seek but rarely achieve because they have neverbefore identified the factors that produce sustained trans-formational change.

55

HBR029ch3 8/15/01 1:35 PM Page 55

The authors identify three interventions that will restorecompanies to vital agility and then keep them in goodhealth: incorporating employees fully into the principalbusiness challenges facing the company; leading theorganization in a different way in order to sharpen andmaintain incorporation and constructive stress; and instill-ing mental disciplines that will make people behave dif-ferently and then help them sustain their new behavior.

The authors discovered these basic sources of revital-ization by tracking the change efforts of Sears, Roebuck& Company, Royal Dutch Shell, and the United StatesArmy. The organizations used these interventions to alterthe way their people experienced their own power andidentity, as well as the way they dealt with conflict andlearning. As at Sears, Shell, and the U.S. Army, anymajor shift in those four elements will create a landmarkshift in any organization’s operating state or culture.

M are trying to make afundamental change in the way they operate. For years,they’ve struggled with growing competition by introduc-ing improvements (or at least improvement programs)into every function and process. But the competitivepressures keep on getting worse, the pace of changekeeps accelerating, and companies keep pouring execu-tive energy into the search for ever higher levels of qual-ity, service, and overall business agility. The treadmillmoves faster, companies work harder, results improveslowly or not at all.

The problem is not the programs, some of which haveworked wonders. The problem is that the whole burdenof change typically rests on so few people. In other

56 Pascale, Millemann, and Gioja

HBR029ch3 8/15/01 1:35 PM Page 56

words, the number of people at every level who makecommitted, imaginative contributions to organizationalsuccess is simply too small. More employees need to takea greater interest and a more active role in the business.More of them need to care deeply about success. Compa-nies achieve real agility only when every function, office,strategy, goal, and process—when every person—is ableand eager to rise to every challenge. This type and degreeof fundamental change, commonly called revitalizationor transformation, is what more and more companiesseek but all too rarely achieve.

Surveys confirm that executives have begun to giverevitalization a high priority. With a few notable excep-tions, however, most of their efforts to achieve it havemet with frustration—partly because large organizationshave such a remarkable capacity to resist change of allkinds, and partly because the kind of change beingsought is so much more radical and uncomfortable thananything required by a shift in strategy or process or cor-porate structure. For that matter, corporate revitaliza-tion often includes shifts in strategy or process or struc-ture, but revitalization means a good deal more—itmeans a permanent rekindling of individual creativityand responsibility, a lasting transformation of the com-pany’s internal and external relationships, an honest-to-God change in human behavior on the job. Revitalizationis not incremental change. Its realizable goal is a discon-tinuous shift in organizational capability—a resocializa-tion so thorough that employees feel they are workingfor a different company, a leap in a company’s ability tomeet or exceed industry benchmarks, a jump in bottom-line results.

This kind of sustained organizational renewal wouldnot be easy even if companies had a reliable road map to

Changing the Way We Change 57

HBR029ch3 8/15/01 1:35 PM Page 57

make the journey a reasonable bet. As it is, most ofwhat’s been written about transformational change iseither too conceptual and therefore too impractical, tooinspirational and therefore too vague, or too companyspecific and therefore too hard to apply to one’s own sit-uation. We have been inept at transforming troubledorganizations—or even at maintaining the vitality ofhealthy ones—because we have never before identifiedthe factors that produce sustainable revitalization.

In essence, there are three concrete interventions thatwill restore companies to vital agility and then keepthem in good health: incorporating employees fully intothe process of dealing with business challenges, leadingfrom a different place so as to sharpen and maintainemployee involvement and constructive stress, andinstilling mental disciplines that will make people behavedifferently and then help them sustain their new behav-ior into the future. Done properly, these three interven-tions will create a landmark shift in an organization’soperating state or culture by significantly altering theway people experience their own power and identity andthe way they deal with conflict and learning.

We discovered these sources of revitalization bytracking the change efforts, in good times and bad, ofthree of the world’s largest organizations: Sears, Roebuck

& Company, Royal DutchShell, and the United StatesArmy. Sears (with $36 bil-lion in revenues and310,000 employees), Shell(with $100 billion in rev-enues and 110,000 employ-

ees), and the U.S. Army (with a $62 billion operating bud-get, 750,000 civilian and active-duty employees, and

Like physicians appraising their patients’health, managers can evaluate the vital signs oforganizational vigor.


HBR029ch3 8/15/01 1:35 PM Page 58

another 550,000 in the Army Reserve) share traits of size,geographical dispersion, and managerial complexity.They also share the distinction of having beaten theodds. All three have survived for 100 years or more andhave retained their essential identity—they have beenneither swallowed up by others nor disaggregated intofragments.

The events that triggered transformation efforts atSears, Shell, and the army were quite different. In allthree organizations, however, the 800-pound gorilla thatimpaired performance and stifled change was culture.The trouble is, there are as many different definitions ofculture as there are articles on change management, andnone of them give us much help in telling us how, oreven what, to fix. Nevertheless, in our study of whatmight loosely be called culture at Sears, Shell, and thearmy, we found four distinct indicators that are highlypredictive of performance in both good times and bad.These four indicators can serve managers in much thesame way that vital signs serve physicians in appraisingthe health of the human body.

This analogy of vital signs is important. The reason somany early forms of healing failed was that practitionerswere treating only the most obvious symptoms of somelarger malfunctioning system they knew little or nothingabout. Gradually, however, medical science identifiedthese invisible systems, figured out how they worked,studied their interdependencies, and learned to pay closeattention to key indicators of a patient’s physical well-being. Today physicians begin an examination by check-ing these vital signs—pulse, blood pressure, pulmonaryfunction, reflexes—to form a general but still fairly accu-rate impression of how each complex subsystem and theorganism as a whole are faring.


HBR029ch3 8/15/01 1:35 PM Page 59

Organizations have similar systems and symptoma-tologies. Their vital signs reveal a great deal about theiroverall health and adaptability, and about the strengthand vigor of their functional systems. The four vital signswe identified at Sears, Shell, and the army give us a work-ing definition of culture and tell us most of what we needto know about the operating state of any company:

Power. Do employees believe they can affect organi-zational performance? Do they believe they have thepower to make things happen?

Identity. Do individuals identify rather narrowly withtheir professions, working teams, or functional units,or do they identify with the organization as a whole?

Conflict. How do members of the organization handleconflict? Do they smooth problems over, or do theyconfront and resolve them?

Learning. How does the organization learn? Howdoes it deal with new ideas?

Organizational Drift

As a result of age, size, or competitive intensity, mostorganizations exhibit a deterioration in vital signs that isinconsistent with—in fact, often destructive to—theirambitions and purposes.

The members of start-up organizations have a senseof individual and collective power; they feel they canmake a big difference in the pursuit of the goals they allshare. Employees identify with the enterprise as a whole;alignment and informal teamwork are commonplace.When conflicts occur, people handle them directly andalmost never allow them to interfere with getting things


HBR029ch3 8/15/01 1:35 PM Page 60

done. The whole organization is open to learning; trialand error are the norm.

As organizations grow older and larger, however, thevigor of these four vital signs deteriorates. Instead ofpower, people often develop a sense of resignation inresponse to seemingly insurmountable obstacles or tolack of support from their superiors in the daily hassle ofgetting things done. As organizations become more com-plicated and demanding, people strive to carve out pri-vate patches of turf where they can exercise responsibil-ity, protect themselves, and keep the world at bay. Whenit comes to their identity, therefore, employees lose theirsense of teamwork and alignment with the entire enter-prise and begin to seek the safety of their particular pro-fession, union, function, team, or location. People inmature organizations tend to avoid conflict for fear ofblame or of having someone take their disagreement per-sonally. Alternatively, they may take part in a successionof routine collisions that lead to stalemate rather thanresolution. As for learning, larger and older organiza-tions tend to be less receptive to new ideas than theiryounger counterparts. In place of inquiry and experi-mentation, ideas get studied to death in hopes of ferret-ing out every possible weakness before making a com-mitment. The precondition for action is certainknowledge.

The Sears story is a useful illustration of a companyculture’s natural drift away from good health. It alsoillustrates one CEO’s well-orchestrated but ultimatelyineffective efforts to reverse the drift and another CEO’sremarkable success at breathing new life into theenterprise.

Ed Brennan’s 12-year tenure at Sears’s helm cannot be faulted for a lack of intelligence, energy, or good


HBR029ch3 8/15/01 1:35 PM Page 61

intentions. He put the Sears Tower up for sale, slimmeddown headquarters, and moved the central organizationto an open campus in Chicago’s suburbs. He called for anend to Sears’s tradition of insisting that customers useonly Sears credit cards. He launched Brand Central,which offered for the first time such non-Sears appli-ances as GE, Maytag, and Panasonic. He diversified intofinancial services through the acquisition of Dean Witter(for $607 million) and Coldwell Banker (for $202 million),

and he invested $1 billionto launch the DiscoverCard. (When he spun offthese assets in 1993, Sears’smarket capitalization hadrisen to $36 billion from $8

billion when he took the job, and the cumulative profitfrom financial services had accounted for approximatelytwo-thirds of Sears’s consolidated earnings for the pre-ceding five years.) Brennan also did his best to rebuildSears as a retail store. He reduced employment by 48,000jobs, simplified logistics, moved into women’s apparel,took steps to streamline the buying organization, and piloted new formats such as stand-alone auto-motive outlets, as well as home-furnishing and home-improvement stores. But it was Brennan’s successor,Arthur Martinez, who became famous for revitalizing the retail side of Sears.

Where Brennan fell down—in contrast to Martinez—was in failing to grapple with the Sears culture; that is,Brennan failed to address the deterioration of the com-pany’s vital signs. Unlike Martinez, he never quite cameto terms with the insight that culture was as strategic ashis product and market initiatives were, and that fixingthe vital signs would go a long way toward fixing thecompany.

CEO Martinez saw thatSears’s culture was as strategic as its product and market initiatives.


HBR029ch3 8/15/01 1:35 PM Page 62

When Martinez took over in 1992, few employees hadany sense of power; most felt nothing but resignation. Asone regional manager put it, “It was a company of ‘saluteand obey.’ Directives came from above, and we did ourbest to follow them. There was no maneuvering room tomake sensible market decisions. As bad press began inthe 1970s, there was widespread depression and anunwillingness to admit at cocktail parties that weworked at Sears. It all seemed so big and complex andout of control. We felt defeated and powerless.”

In terms of identity, Sears had drifted a long way fromthe vibrancy that prevailed from its founding in 1880through 1956, when General Robert Wood retired as itsfourth CEO. For much of that period, the central buyingstaff and equally strong territory managers kept eachother honest through a system of checks and balances.Beginning in the mid-1950s, however, a succession ofcaretaker CEOs allowed this precise tension between thefield and the home office to degenerate into an empty rit-ual. Territory managers ran their stores like baronies andstonewalled strategic direction from above. This tilttoward regional fragmentation and a more local identityunquestionably contributed to Sears’s inability torespond early to the threats posed by Wal-Mart andToys-R-Us. “Too small to worry about” or “Not a problemin my region” were the typical reactions.

Brennan tried to correct the excesses of decentraliza-tion but pushed the pendulum too far back in the oppo-site direction. He eliminated the position of territorymanager along with most other echelons in the regionalhierarchy. Once-powerful store managers were relegatedto “keeping the lights on and the doors open,” as onestore manager put it. “As a result, our knowledge at thefingertips was lost. Executive management ushered in anera of drive-by merchandising. Experts from headquar-


HBR029ch3 8/15/01 1:35 PM Page 63

ters would visit a store three times a month and wouldbelieve they understood your local market better thanyou did.”

Under enormous pressure to meet their performancetargets and threatened by further layoffs, store manage-ment teams hunkered down and concentrated on theirown turf. As a consequence, the stores became a mer-chandising hodgepodge, and poor service and frequentout-of-stock conditions alienated customers. Many ofBrennan’s efforts to achieve sweeping change snagged inthe concertina wire of the stores’ defensive perimeters.Brennan was trying to build a companywide identity byedict. Predictably, the center did not hold, and the effortfailed. In the fallout, identity fragmented more than everas people everywhere in the company looked out forthemselves.

The third of Sears’s vital signs to show serious deteri-oration was its capacity for constructive conflict. Thecompany’s initial operating model was built on a vibranttension between home and regional offices. This strugglebetween policy from headquarters and inventive execu-tion at the stores was mediated by a succession of fourvery strong and accessible CEOs whose careers spannedthe first 76 years of Sears’s history. These men were notthreatened by conflict; they encouraged and harnessed it.But by the time Ed Brennan succeeded to the chairman-ship, the former mix of initiatives and controls had longsince given way to compliance and acquiescence. Push-ing back or resisting directives meant “not being a teamplayer.”

As for learning, Sears had drifted to the most extremecondition of denial and complacency. Negative reportsby business analysts beginning in 1974 were interpretedat Sears as bad journalism and unfair treatment. In 1990,


HBR029ch3 8/15/01 1:35 PM Page 64

when Sears was accused of widespread dishonesty in itsautomotive repairs (a sting operation by the State of Cal-ifornia Consumer Affairs Department found fraud in85% of its visits to Sears Auto Centers, and 44 states sub-sequently filed suit), Brennan’s first line of defense was

to deny all allegations and tostonewall. An even moretroubling example: Searsbegan including Wal-Martamong its competitor bench-marks only in 1992, by whichtime Wal-Mart was 60%larger than Sears. Sears’s

strongly inbred culture was deeply implicated in thiscapacity for denial. Throughout Brennan’s tenure, onlyone of 100 top executives had a non-Sears career back-ground. “Sears is different” or “We tried that once and itdidn’t work” were frequent responses to new ideas.

Incorporating Employees

Sears, Shell, and the U.S. Army are currently engaged inefforts to revitalize their organizations. All three aredoing their best to transform the way their people expe-rience power, identity, conflict, and learning. All three, inone fashion or another, are using the same three inter-ventions to achieve this improvement in their vital signs.

The first intervention is to incorporate employees intothe activity of the organization. This is not the same ascommunicating or motivating or rolling out planshatched at the top. It is resocialization. It means engag-ing employees as meaningful contributors (not justdoers) in the principal challenges facing the enterprise. Itmeans seeing employees as volunteers who decide each

The first step toward restoring organizational vitality is to engage every employee in thecompany’s principalchallenges.


HBR029ch3 8/15/01 1:35 PM Page 65

day whether or not to contribute the extra ounce of dis-cretionary energy that will differentiate the enterprisefrom its rivals. Although incorporation shares DNA withsuch familiar ideas as consensus management, employeeinvolvement, and self-managed teams, it is somethingmore. Its distinct properties include the use of concrete,pressing business problems to generate a sense ofurgency; the cascading involvement of every employeebeginning at the very top of the enterprise and continu-ing down through the ranks; and the generation of initia-tives conceived and staffed by employees across hierar-chy and function.

We can see one leader’s efforts to reverse driftthrough incorporation in the turnaround of ShellMalaysia. Its British chairman, Chris Knight, had thebenefit of three career rotations in Malaysia prior to hisappointment as chairman. When he arrived in 1992, he saw that the organization was in trouble. The com-pany was overstaffed; traditional revenues from oil andgas were in decline; service standards with wholesalecustomers were in disarray; and the once-dormantgovernment-owned oil company, Petronas, had becomean aggressive competitor in the vehicle-fuel market.

Knight wanted to build a much more agile and lesscostly enterprise, but he had watched several predeces-sors try and fail to alter Shell Malaysia’s vital signs. Mostemployees felt that as the largest private oil company inthe country, Shell should try not to make waves. Thisultraconservative philosophy led employees to avoid anydeviation from usual practice and stifled in the cradleany impulse to use their power of initiative. Within acocoon of comfortable oligopoly, their identity waslocated in small, defensible silos. Refining quarreled withtransport, and everyone fought the crazy ideas that came


HBR029ch3 8/15/01 1:35 PM Page 66

from marketing and sales, but all these conflicts weredistinctly muted. Malaysian employees are from culturessensitive to saving face and therefore tend to approachimpasses by highly circuitous routes. “Smooth andavoid” was the norm. Finally, there was little learning.Knight observed a frustrating lack of concern, even ofcuriosity, when competitor Caltex gobbled up 10% ofdomestic market share.

For more than a year, Knight tried to achieve authen-tic alignment among his eight-person executive team.Somehow, the goal always eluded his grasp. In exaspera-tion, he scheduled an incorporation event in Kuching,Borneo, and asked all 260 of Shell’s senior and midlevelmanagers to attend.

The leader of a middle-management strategic-initiative team kicked off the two-and-a-half-day meet-ing with a brief presentation of two key proposals aimedat repositioning Shell and regaining competitive advan-tage. The first proposal envisioned a daring partnershipwith Shell’s biggest competitor, Petronas, in order toengage in joint procurement, thus lowering costs forboth companies and putting their competitors at a dis-advantage. The second proposal was to streamline andimprove Shell’s ragged relationships with its 3,000 fran-chised service stations by creating a single point of con-tact in a customer service center.

Assembled in small groups, the managers were thenasked to identify the soft spots in these strategic propos-als. When the entire assembly reconvened, some groupssuggested improvements from the floor, but on balancethere was general agreement with the proposals. Thenext step was an organizational audit. Each of severallarge teams of participants took one facet of the com-pany—strategy, structure, systems—and described how


HBR029ch3 8/15/01 1:35 PM Page 67

it affected current performance and the impact it mighthave on the two proposals. When these analyses wereshared in plenary session, it was evident to most peoplethat Shell’s operating practices would seriously compro-mise the new initiatives. Over the course of the meeting,many of Knight’s management cadre became aware ofthe emerging competitive pressures affecting the com-pany and were mobilized to take part in developing aresponse. Such mobilization is the aim of any well-designed incorporation process.

Just below the surface of this off-site meeting, anotherdevelopment was taking place. As lower-level managersgained firsthand knowledge of business priorities andsaw where the chairman wished to take the company,the vast majority of them bought into the plan, which leftobstructionist senior managers isolated and exposed.One senior British expatriate, recognizing that his handhad been called, chose the final ten minutes of the meet-ing to air his differences with the chairman publicly.Knight dismissed him 48 hours later—a firing heardround the world of Shell, where this sort of thing wasnever done.

But the firing raises an obvious question: How doesdismissal for disagreement fit together with the notion ofencouraging constructive conflict? Knight’s position wasthat he fired the man not for disagreeing but for neverdisagreeing in the previous 13 months of high-level dis-cussions or at any time during the meeting except in theconcluding minutes. Moreover, most Shell employees—at least those in Malaysia—accepted this explanation.Rather than create a fear of openness, the termination ofan executive widely seen as an opponent of change wasregarded as a defining moment in the progress of thebroader involvement and deeper commitment that isincorporation.


HBR029ch3 8/15/01 1:35 PM Page 68

Incorporation doesn’t begin and end with one off-sitemeeting, however uplifting it might be. Knight’s nextmove was to sponsor one-day events called valentines, aname and concept that he borrowed from Ford. In theseexercises, gatherings of 100 salaried and hourly employ-ees split into smaller groups of peers from each of the

major functional unitswithin the downstreamorganization—refining,logistics, engineering, cus-tomer service, accounting,and so forth. At issue was

the new concept proposed in the second strategic initia-tive outlined in Kuching—the customer service center.Knight’s goal was, first, to give customers a single pointof contact with Shell through a toll-free number and, sec-ond, to empower the customer service center to breaklogjams and satisfy customer needs. The second of thesetwin objectives was the wolf in sheep’s clothing. It is easyenough to recruit an around-the-clock staff of operatorsto cover a toll-free number. It is quite another matter toshift organizational power so radically that customer-service-center representatives will be able to break dead-locks and redeploy resources. This is the stuff over whichorganizational blood is spilled—and a challenge that didnot play into the historical strengths of Shell’s down-stream functions.

The valentines exercise is a vehicle for conflict resolu-tion. Each functional team is required to write a succinctdescription of its grievances with any of the other teamsin the room, pinpointing what it does to inhibit produc-tivity and what is likely to get in the way of a successfulcustomer-service center. When each group has received,say, half a dozen of these valentines, its members aregiven time to sift and discuss them, and then to select

In the valentines exercise,people air conflicts and struggle toward arobust resolution.


HBR029ch3 8/15/01 1:35 PM Page 69

two issues they think particularly important to resolve.The group then gets two hours to come up with, first, adetailed plan for corrective action that it can implementwithin 60 days; second, the name of a member of its ownteam who will be accountable for delivering the action;and third, the name of a so-called committed partnerfrom the team that sent the valentine who will shareresponsibility for making the new solution work.

Back in plenary session, each person assigned anaction stands and explains the grievance and the pro-posed solution, and names the team’s nominee for com-mitted partner—often the individual seen as most likelyto sabotage the proposal and therefore the person mostessential to its success. The committed partner thenstands, and a fascinating negotiation unfolds. Tensionmounts, and the room falls silent. With coaching fromthe facilitator, the two principals air conflicts andexpress their deep-seated distrust of each other’smotives. A robust solution is the usual result.

Making good use of these and other techniques, ShellMalaysia reversed its ten-year drift. It fostered a newlevel of individual power, a new sense of identity with theenterprise as a whole, a new kind of open and productiveconflict, and a new appetite for learning that persists tothis day.

Leading from a Different Place

An organization coming unfrozen under an overload ofexperimentation and new ideas is a terrifying thing fortraditional leaders. Matters seem out of control, which toa degree they are. But as leaders weather this storm, theybegin to undergo a shift in mind-set. From thinking, “I’vegot to stay in control” or “This is too fast,” they develop


HBR029ch3 8/15/01 1:35 PM Page 70

an ability to operate outside their comfort zone andaccept ambiguity and adversity as a part of the design.The second of the three interventions—a new approachto leadership—requires them to establish focus andurgency, maintain healthy levels of stress, and not feelcompelled to come to the rescue with a lot of answers.They learn to stay the course until guerrilla leaders atlower levels come forward with initiatives that addressthe company’s shortcomings.1

Arthur Martinez did precisely all these things atSears. And from the very beginning, he did one impor-tant thing that Brennan had not done: he began tellingthe truth. For seven successive years, retail executives atSears had lied to themselves. They set annual goals andcame back at year’s end below plan. The targets were setlower each time around, but they were never low enough.Market surveys showed Sears perilously close to breakingits last remaining links with its retail customers. It wasthe Sears credit card, delivering 70% of the profits, thatwas carrying the retail group. All of this was painful toface. Martinez held up the mirror.

To generate a sense of urgency, Martinez set difficultgoals. Within two years, Sears would quadruple its mar-gins to achieve industry parity, reverse its loss of marketshare, and improve customer satisfaction by 15%. Thencame the hard part. Like Sears under Brennan, mostorganizations are submerged in their numbing but famil-iar lethargy, their somnambulant operating state. It’s likebeing stirred from your dreams by a strange noise in thenight: in the fog of semiconsciousness, one part of youstruggles to focus on whether it’s an intruder or the cat;but another part resists the possibility of bad news andstruggles to go back to sleep. Similarly, people in organi-zations resist undertakings that would pull them from


HBR029ch3 8/15/01 1:35 PM Page 71

their familiar world. When a leader raises an issue andgenerates urgency around it, the guaranteed first line ofdefense is for the organization to turn back to the leaderfor an answer. “We need a plan . . . more direction . . .more resources” are the words to this predictable refrain.Many leaders take the bait. Martinez did not. He refusedto give his team of top-level managers the answers, andthe authenticity of his refusal was powerful. He didn’thave the answers. No one did. Sears’s management hadto create the answers on the basis of what Martinez didprovide—which was truth, urgency, and enough produc-tive stress to alter thinking and behavior.

Leading from a different place always requires reso-cialization of the kind Martinez achieved at Sears.Nowhere is the transformational power of resocializationmore evident than at three highly unusual U.S. Armytraining centers—at Fort Irwin, California; Fort Polk,Louisiana; and Hoenfelds, Germany. In fact, the trainingis sufficiently remarkable to have been studied by thechief education officers at Shell, Sears, Motorola, and GE,and by senior delegations from every country in WesternEurope, Russia, and most nations of Asia, Latin America,and the Middle East. Perfected over the past 15 years, the training is widely recognized to have almost single-handedly transformed the army, the largest employer inthe United States.

Over a grueling two-week period, an entire organiza-tional unit of 3,000 to 4,000 people goes head-to-headwith a competitor of like size in a simulation so realisticthat no participant comes away unscathed. The exerciseoften alters forever the way executives—in this case,army officers—lead. Critical to its impact is a cadre of600 instructors, one assigned to every person withleadership or supervisory responsibilities. These


HBR029ch3 8/15/01 1:35 PM Page 72

observer/controllers, as they are called, shadow theircounterparts through day after 18-hour day of intenseactivity. They provide personal coaching and facilitate anonhierarchical team debriefing called an After ActionReview (AAR), in which participants struggle to under-stand what went wrong and how to correct their short-comings. These AARs are in fact the focal point of anorganizational exercise that can range across 650,000acres (at Fort Irwin in the Mojave Desert) and cost $1million a day.

For many, the juxtaposition of U.S. Army with wordslike revitalization, experimentation, and nonhierarchicalamounts to a contradiction in terms. But that view is outof date. According to General Gordon R. Sullivan, thearmy’s recently retired chief of staff, “The paradox of war

in the Information Age isone of managing massiveamounts of informationand resisting the tempta-tion to overcontrol it. Thecompetitive advantage isnullified when you try to

run decisions up and down the chain of command. Allplatoons and tank crews have real-time information onwhat is going on around them, the location of the enemy,and the nature and targeting of the enemy’s weaponssystem. Once the commander’s intent is understood,decisions must be devolved to the lowest possible level toallow these frontline soldiers to exploit the opportunitiesthat develop.”

A number of factors have contributed to the army’sextraordinary, sustained transformation, includinghigher-quality soldiers, one outcome of a volunteerarmy. But inside and outside observers agree that the

Leading from a different place means resisting the temptation to provide the answers. Solutions must come from the ranks.


HBR029ch3 8/15/01 1:35 PM Page 73

National Training Command (NTC) has been the cru-cible in which it has all come together. Since the NTCwas established, the army’s more than half a millionmen and women in uniform have rotated through itsprograms several times—most upper-, mid-, and lower-level officers and NCOs, five times. As one officer put it,“The NTC experience leaves no room for debate. Dayafter day, you are confronted with the hard evidence ofdiscrepancies between intentions and faulty execution,between what you wanted the enemy to do and what heactually did.”

Leading from a different place requires great resolveboth to stay the course and to resist the temptation toprovide the answer. The solutions, and the commitmentto deliver on them, must come from the ranks. Leaders

must maintain the pres-sure until followers seethat they are going to haveto make things happen,until guerrilla leaders stepforward and begin toengage in leaderlike acts.

Not everyone is a guerrilla leader, but sustained stresswill eventually produce enough such leaders to beginshifting the tide of vital signs.

Leading from a different place also entails a transfor-mation in the operating state of leaders themselves. Theybecome a microcosm of the shift in vital signs that theywant to see in their organizations. From resigning them-selves to the limits of their power to make things happen(and to the implausibility of expecting middle managersto help), they move toward the possibility of genuinelydistributed intelligence; from taking on an identity as theperson in charge, they become clearinghouses for thedifferent ways an enrolled organization handles its

Leading from a different place also requires leaders to stand squarely in the zone of discomfort and ambiguity.


HBR029ch3 8/15/01 1:35 PM Page 74

responsibilities; from avoiding straight talk, they developan ability to handle and even encourage constructiveconflict; from assuming that they must provide adetailed road map for the journey, they begin to acceptlearning as a form of inquiry in action. Leaders mustplace themselves squarely in the zone of discomfort andlearn to tolerate ambiguity. We are all much more likelyto act our way into a new way of thinking than to thinkour way into a new way of acting, and that is the essenceof leading from a different place.

Instilling Mental Disciplines

We know that when incorporation slackens or van-ishes—as it did at Sears for a long time before Martinezor in the army before and during Vietnam—stagnationand entropy are almost invariably the results. We haveseen at Sears, Shell, and the U.S. Army that incorporationcombined with a different type of leadership was able toreverse an organization’s drift and restore its culturalvitality. But if an organization is to change the way itspeople think and act and interact, and if this resocializa-tion is not to evaporate the moment financial resultsimprove and people start to believe the worst is over,then people must internalize a set of principles or disci-plines that shape their reactions and govern their behav-ior. Disciplines of this kind might also be called enduringsocial patterns, but they are a good deal more thanunconscious habits. Habits are automatic and thereforemindless. Disciplines are mindful. We can see thesedisciplines at work in the After Action Review, whichconstitutes the heart of the NTC experience.

Each afternoon, the commander of the brigade under-going training receives an assignment, such as “pene-trate enemy defenses” or “defend your sector against a


HBR029ch3 8/15/01 1:35 PM Page 75

superior force.” Inside crowded command tents, 30 to 40staff officers and senior fighting-unit commanders studythe situation and endeavor to hammer out a winningstrategy. Later that afternoon, this strategy begins to fil-ter out to 3,000 soldiers dispersed across many squaremiles of rugged terrain. Tank crews and platoons arebriefed, minefields laid, artillery and helicopters coordi-nated, reconnaissance initiated. Commencing at mid-night, both friendly and enemy probes get under way.

By dawn, the day’s battle is in full swing. The “enemy”(the 11th Armored Cavalry Regiment) is permanentlystationed at Fort Irwin. It knows the terrain, behavesunpredictably, and almost always devastates the unit intraining. And all the action is recorded. Perched onmountaintops, powerful video cameras zoom in on thehot spots. An elaborate laser-based technology preciselytracks when and where each weapon is fired, electroni-cally disabling any fighting unit that is hit. Audiotapesrecord communication and confusion over the voice net-work. By 11 a.m., the outcome has been decided, andwithin 90 minutes, the observer/controllers have pulledeach combat team together near terrain that has beenpivotal in its piece of the battle.

Let us take a closer look at an AAR in progress.2 Acompany team of two platoons with two tanks, four ar-mored personnel carriers, and an HMMV (the modernversion of a jeep) have pulled into a tight circle underthe shade of a desert outcropping. The crews lean backagainst tank treads, a flip chart slung over the HMMVantenna. The fighting is in its fifth day. Exhaustion isevident. The observer/controller has created a sandtable on the ground, a miniature of the terrain in whichthis unit was annihilated in the day’s battle. He asks agunnery sergeant to come forward, position the com-


HBR029ch3 8/15/01 1:35 PM Page 76

pany’s armor on the sand table, and explain the unit’smission.

Sergeant: Our overall mission was to destroy theenemy at objective K-2.

Observer/Controller: Why was this important? Whatwas your tank’s particular role in all this?

Sergeant: I’m not sure.

Observer/Controller: Can anyone help?

A trickle of comments gradually builds into a flood ofdiscussion. It begins to appear that only the lieutenant incharge understood the mission. There had been no coor-dination of individual tanks and vehicles, and none hadbeen given a particular sector in which to concentrate itsfire. No one had understood that the unit’s main taskwas to drive the enemy column away from a weak pointin the defenses and into a zone where it would be withinrange of friendly tanks and artillery.

Key lessons for the next day are recorded on the flipchart. The soldiers all come away with a picture of whatthey were involved in but could not see. Each soldier hascontributed to a composite grasp of the engagement,supplemented by video clips and hard data from theobserver/controller. Day after day, particular themes arereinforced: all members of the unit must understand thebig picture; they all need to think; they must always putthemselves in the shoes of an uncooperative enemy; theymust prepare to the point that surprise will no longersurprise them; they must set aside hierarchy, exerciseself-criticism, work as a team.

“The After Action Review has democratized the army,”says Brigadier General William S. Wallace, current com-mander of the NTC. “It has instilled a discipline of relent-


HBR029ch3 8/15/01 1:35 PM Page 77

lessly questioning everything we do. Above all, it has reso-cialized three generations of officers to move away from acommand-and-control style of leadership to one thattakes advantage of distributed intelligence. It has taughtus never to become too wedded to our script for combatand to remain versatile enough to exploit the brokenplays that inevitably develop in the confusion of battle.”

The success of the NTC experience and the AfterAction Review is the result of carefully designed impera-tives that can be applied in any organization or corpora-tion. First, take a team of people who must work togetheracross functions and hierarchies and immerse it in a pro-longed, intense learning experience. Have the team takeon a very tough project or a very tough competitor. Underthe right conditions, stress and exhaustion will unfreezeold patterns of behavior and create an opening for newunderstanding and behavior to take root. Second, in orderto eliminate subjectivity and debate, collect hard data onwhat has transpired. Let the data, not the trainers, pointthe finger. Third, utilize highly skilled facilitators whohave a deep knowledge of what they are observing. Nevercriticize. Use Socratic questioning to evoke self-discovery.Fourth, do not evaluate performance. The experience isnot about success or failure. It is about how much eachindividual can learn. Make it safe to learn.

There are seven disciplines embedded in the AfterAction Review, and all seven are as relevant in businessas they are in combat.

1. Build an intricate understanding of the business.An organization’s members do best when line-of-sightunderstanding bridges the gap between overall strategyand individual performance. This is harder than it looks.On the one hand, troops need to understand the princi-pal aims of each engagement (“move to establish contact


HBR029ch3 8/15/01 1:35 PM Page 78

but don’t precipitate an all-out fight” or “block theenemy at this line but don’t commit to a counterattack”)and how it fits into the larger strategic context. On theother hand, soldiers need solid individual skills. Bothrequirements are essential. The idea is to prevent sol-diers from behaving like automatons. They are not theresimply to obey orders but to apply their skills and intelli-gence to a larger goal.

The first requirement—conveying the big picture tothe small unit—is easy to overlook in the heat of prepar-ing for battle. In the AAR close-up above, we saw howthe lieutenant commanding the armored unit hadneglected to communicate the big picture and how hismen then failed to achieve a goal of which they wereunaware. To carry out the second requirement, develop-ing individual areas of expertise, the army has borroweda concept from the total quality movement and has dis-tilled all the facets of a military action down to three: thekey tasks involved, the conditions under which each taskmay need to be performed, and the acceptable standardsfor success. (For example, at a range of 2,000 yards, hit anenemy tank moving at 20 miles per hour over uneven ter-rain at night with an 80% success rate.)

Sears has shown an exemplary grasp of this discipline.To convey the larger strategic picture to every employee,the company uses learning maps—large murals with

elaborate legends on theborders—to communi-cate essential businessconditions to smallgroups of employeesworking with a facilita-tor. One map takes peo-

ple through the shifts in the competitive environmentfrom 1950 to 1990. Another map, laid out like a game,

Frank exchanges are not likely to occur if enlisted men are holding back out of deference to theirsuperiors.


HBR029ch3 8/15/01 1:35 PM Page 79

asks employees to place bets on the sources and uses offunds as they flow from customers’ wallets to the bottomline. Sears then asks its employees to use what they’veabsorbed from the learning maps to come up with a listof three or four highly practical actions that can be takenimmediately at the store level to correct deficiencies andimprove customer service.

Sears anchors the proficiency side of this disciplinewith training to improve its interface with customers,then adds performance measures that focus attention onindividual and team performance with respect to cus-tomer satisfaction. Together, these initiatives enableemployees to perform to high standards and to under-stand how they each contribute to Sears’s success.

2. Encourage uncompromising straight talk. TheAAR is predicated on a frank exchange among soldiers asthey sort through the confusion of battle and figure outwhere things went wrong. Such an exchange will notoccur if people are showing deference to their superiors orholding back for fear of hurting someone’s feelings. As wenoted earlier, observer/controllers are skilled at usingobjective data to point the finger—fostering healthy give-and-take and creating a safe environment for candor.

Sears practiced this discipline from the top down(Martinez helped his top-level managers confront thetruth about Sears’s past performance) and from the bot-tom up (town hall meetings cultivated a new and muchmore straightforward style of communication). We alsosaw Shell Malaysia emphasize this discipline with itsvalentines exercise.

3. Manage from the future. Hardship for its own sakeis clearly not the army’s intention, but attaining excel-


HBR029ch3 8/15/01 1:35 PM Page 80

lence can be painful. Be All That You Can Be is morethan the army’s recruiting slogan. It challenges every ele-ment of the institution—from the private soldier to thelogistics command—to stretch itself. Being all you can beis not a destination to be reached but a mind-set to man-age from.

Organizations often “use up” their future, and that isprecisely what happened to the U.S. Army after the high-water mark of World War II, to Sears after GeneralWood’s retirement, and to Shell in the 1980s. Once themembers of an organization believe they have reachedthe future, they begin to codify their past successes. Driftand loss of vitality follow “winning formulas” of this kindjust as surely as night follows day.

The most essential aspect of managing from thefuture is to alter the institution’s point of view. We alltend to look toward the future as a distant goal. By con-trast, this discipline means internalizing some futuregoal so the institution can plant its feet in that futureand manage the present from there. At Shell Malaysia,

Knight inherited a com-pany that had used up itsfuture, a company contentto keep a low profile as ittried to avoid further mar-ket-share losses to Caltex,Mobil, and Petronas.

Knight shifted this mind-set entirely, asserting that thefuture of the industry was regional, not national. Heinsisted that Shell and Petronas needed to join forcesand make Malaysia the dominant low-cost player inSoutheast Asia. Once this perspective was accepted as avalid view of the future, a stream of beneficial resultsflowed from it for both companies.

Observer/controllers hammer on the benefits of controlled failure until soldiers embrace setbacks as windows to learning.


HBR029ch3 8/15/01 1:35 PM Page 81

4. Harness setbacks. NTC participants know from theoutset that they are fighting an enemy far tougher thanany they are likely to meet in the field. Observer/con-trollers remind them daily that their maneuvers are notabout winning but about learning. Harnessing setbacksis a matter of recontextualizing failure, treating break-downs as breakthroughs, seeing defeat as opportunity.But this requires considerably more self-discipline thanmost managers realize. Human beings are hardwired toreact adversely to mistakes by blaming themselves (guiltor shame), others (finger-pointing), or bad luck (resigna-tion and fatalism). Day after day, observer/controllershammer on the benefits of controlled failure until everysoldier learns to embrace setbacks as windows tolearning.

This discipline has been directly applied at Sears,where Gus Pagonis (one in a long line of increasinglysought-after U.S. Army generals who have landed topcorporate posts) heads Sears’s far-flung logistics empire.Pagonis has brought the entire AAR process directly toSears. Daily sessions of 10 to 12 employees representingevery level from warehouse to headquarters scrutinize24-hour updates on late or wrong shipments and chipaway at corrective action.

5. Promote inventive accountability. The tasks, con-ditions, and standards in the first discipline create thebenchmarks of acceptable performance, and soldiers aretrained to meet or exceed these benchmarks so that theirunits can count on them in combat. But there is more toit than that. Close battles are won by exploiting theenemy’s broken plays. Mastery of a combat assignmentrequires not just replicable skills but also the capacity to


HBR029ch3 8/15/01 1:35 PM Page 82

improvise. Observer/controllers single out and rewardcreative acts of initiative that are built on a solid plat-form of proficiency.

The new emphasis on “the softer side of Sears” hasbrought the company into competition with Nordstrom,which, according to Martinez, sets the world standard instriking the proper tension between improvisation andaccountability. Nordstrom encourages inventivenesswith the motto Respond to Unreasonable CustomerRequests (for example, delivering an over-the-phone pur-chase to a frantic customer at the airport who is about tocatch a plane). Salespeople keep scrapbooks of theirheroics in providing exceptional service, and these hero-ics figure into promotions and storewide recognition. Onthe accountability side, each department tracks the salesper hour of each salesperson and posts the informationpublicly every two weeks—from the top of the list(worst) to the bottom (best). A sales associate unable tomeet a threshold level of sales on a three-month rollingaverage is dismissed—an infrequent occurrence, sincetopping the list several times in a row leads most poorperformers to move on of their own free will.

6. Understand the quid pro quo. Organizationalagility and the disciplines that sustain it make enormousdemands on people. Organizations must make sure thattheir members receive commensurate returns. Onceupon a time, corporations were like ocean liners. Anyonefortunate enough to secure a berth cruised right throughto disembarkation at retirement. In return for loyalty,sacrifice, and the occasional aggravating boss, employeesat Sears, Shell, and the army, among others, enjoyedimplicit or explicit job security.


HBR029ch3 8/15/01 1:35 PM Page 83

We have now witnessed a decade of continuous jobattrition in which companies have downsized, delayered,

reengineered, and out-sourced. From 1980 to1996, Sears has laid offmore than 100,000employees. The U.S.Army has reduced its

ranks by 300,000 soldiers from a high of 1.2 million dur-ing the Gulf War. Worldwide, Shell has cut 150,000 jobssince 1980.

Understanding the quid pro quo is a demandingdiscipline. A genuinely transformational employmentcontract has four levels—three more than the reward-and-recognition that was once considered adequate.The second level is employability—the training andskills that enhance people’s marketability. Valuable asthis element of the contract may be over the long run,it is nevertheless overrated as an incentive. Enhancedemployability will inspire no one to offer the kind ofdeep, creative commitment and enthusiasm that com-panies struggling for revitalization so badly need. Em-ployee involvement at that level cannot be bought orenticed, and it is not likely to emerge naturally from theindividualistic, transactional employment contractsthat are typical for many kinds of credentialed expertsand specialists.

It takes more than compensation and employabilityto produce transformational participation. It also takes asense of meaning in the work strong enough to generateintrinsic satisfaction. And finally, employees mustunderstand where the enterprise is going and have somesay in shaping its destiny. Shell, Sears, and the army areall wrestling with these four components of the quid pro

Employees must understand where the enterprise is going and have some say in its destiny.


HBR029ch3 8/15/01 1:35 PM Page 84

quo. In the army, the AAR is the engine of a powerfullearning and resocialization experience, driven in part bypeople’s clear perception that defending their country isimportant work. Senior officers take part in more dra-matic changes of perception than their juniors; but eventhe lowest-ranking soldier has a hand, day after day, inaltering the army’s culture and, ultimately, its destiny.

7. Create relentless discomfort with the status quo.The After Action Review is based on the notion that indi-viduals can improve—in most cases, improve dramati-cally—on everything they do. Observer/controllers con-tinually reinforce the notion that AAR disciplines can beapplied elsewhere to other activities, and a protocol likethe AAR does tend to get under a person’s skin. Soldierscarry the ideas back to their home bases. Once internal-ized, the discipline of relentless discomfort begins toreveal itself in repeated, gnawing questions: How can wedo this still better (faster, cheaper)? Is there a radicalnew approach that we haven’t thought of yet? Day in andday out, throughout the army, the AAR format and disci-plines are employed to critique performance and tomake improvements as soldiers and employees at everylevel begin to see acceptable performance levels as insuf-ficient for sustained vitality.

Sears and Shell struggle to turn episodic attention toimprovement into a vigorous daily discipline. Amongtheir benchmarks is USAA, long a top performer in theinsurance industry. USAA has adopted a practice it calls“painting the bridge”—a reference to the fact that thetask is never complete. (As soon as painters on the GoldenGate or any other large bridge finish the job, it’s time to goback and start over.) In brief, an independent team of 14organizational experts starts at one end of USAA and


HBR029ch3 8/15/01 1:35 PM Page 85

works its way to the other, one unit at a time. Its missionis to work with departmental teams and question every-thing they do. Is the role teams perform necessary? Can itbe streamlined or improved? Can the team be mergedwith another unit? Can it be eliminated? Not surprisingly,people in the company have ambivalent feelings aboutthis once-every-two-year regimen. But it reliably deliversimprovements and, equally important, reinforces USAA’sunending effort to become a better company.

Researchers at the Harvard Business School recentlytracked the impact of change efforts among the Fortune100. Virtually all these companies implemented at leastone change program between 1980 and 1995, but only30% of those initiatives produced an improvement inbottom-line results that exceeded the company’s cost ofcapital, and only 50% led to an improvement in marketshare price. This discouraging result was not for lack oftrying. On average, each of the companies invested $1billion in change programs over the 15-year period.3

Frustration with such results is naturally widespreadbecause the effort and the outcome are so hugely dispro-portionate. Or to be more precise, the effort of some peo-ple in a company is so much greater than the outcome forall. The solution is to focus on the all, to shift the atten-tion from incremental change to the tools that can trans-form the attitudes and behavior of every last employee.

Notes

1. See Ronald A. Heifetz and Donald L. Laurie, “The Work ofLeadership” (HBR January–February 1997).

2. The following description is taken from a video presenta-tion of an After Action Review at Fort Irwin, titled


HBR029ch3 8/15/01 1:35 PM Page 86

Mojavia: In Pursuit of Agility (New York: Marc GersteinAssociates, 1997).

3. Nitin Nohria, “From the M-form to the N-form: TakingStock of Changes in the Large Industrial Corporation”(Harvard Business School Working Paper 96-054).

Originally published in November–December 1997Reprint 97609

This article is based on research and consulting work carried out jointlywith CSC Index.


HBR029ch3 8/15/01 1:35 PM Page 87

Racing for GrowthAn Interview withPerkinElmer’s Greg Summe

Executive Summary

BY THE TIME GREG SUMME joined EG&G in 1998,the company badly needed to shed the weight of pastglories and rediscover the technological innovation thathad been at its heart. An alumnus of McKinsey & Com-pany and a veteran of GE and AlliedSignal, Summe, 43,saw may strengths but no strategic coherence. “Whenyou put the parts together,” he says in this detailed inter-view, “you didn’t get a more valuable whole.”

First as president and COO and later as chairman and CEO, Summe applied a cool rationalism to the com-pany’s strategy and paid close attention to preparing itspeople for a new competitive environment. The result, lessthan three years later, is PerkinElmer, a high-tech darlingwhose stock has more than tripled since Summe’s arrival.

The first task, he says, was establishing more ambi-tious performance goals, specific metrics and rewards,

89

HBR029ch4 8/15/01 1:35 PM Page 89

and more accountability. The company also consoli-dated its 31 businesses into four strategic business units,integrated sales forces, shifted production to the Far East,developed a corporatewide materials-purchasing pro-gram, and raised profit and growth goals. It establishedfour rigorous, corporatewide processes: goal setting todrive strategy; a leadership and organizational review todevelop talent; an annual operating plan to set perfor-mance goals; and a procurement, quality, and productiv-ity program for continuous improvement.

Summe provides details of PerkinElmer’s training initia-tives and describes how the company sought a newcommunity of Wall Street analysts, allowing it to reach abroader base of growth-oriented investors.

If you had to choose a two-word description for the com-pany Greg Summe joined in 1998, it would be “couchpotato.” While other technology companies soared tonew heights, EG&G seemed to be channel surfing from acomfortable old chair, idly wondering from time to timewhat all the commotion was about. Fifty years and some90 acquisitions after its founding in 1947 by three MITprofessors, EG&G had become diversified to the pointof confusion. With most of its business in low-margin tech-nical services for the U.S. government—from programmanagement for the Navy’s nuclear submarine fleet toincinerating the Army’s chemical weapons stockpile inUtah—EG&G badly needed to shed the weight of pastglories and rediscover the technological innovation thathad originally been the heart of the company.

Summe’s task, first as president and COO and, ayear later, as chairman and CEO, was nothing less than

90 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 90

to reinvent the company. An alumnus of McKinsey &Company and a veteran of GE and AlliedSignal, wherehe had run several aerospace and industrial divisions,Summe, 43, is a cool rationalist, a disciplined manager,and the kind of person who would rather head out thedoor for a lunchtime run than amble over to the cafete-ria. His energy and clarity of purpose contrastedsharply with EG&G’s make-do culture.

Just over two and a half years later, Summe leads acompany that has become a hot ticket. Not only doesEG&G have a new name—PerkinElmer—but it has alsobeen revamped from an old-economy dullard into ahigh-tech darling. It has shed ten businesses, completedseven acquisitions, and reorganized its remaining busi-nesses into four distinct business units. Early results arepromising: PerkinElmer’s stock has more than quadru-pled since Summe’s arrival, its operating margin hasnearly doubled, and its organic growth rate has jumpedfrom zero to nearly 12%.

In this interview, conducted at PerkinElmer’s corpo-rate headquarters in Wellesley, Massachusetts, Summedescribes the theory and practice behind that perennialquestion: how do you radically transform an organiza-tion? For PerkinElmer, the answer was to prune its busi-ness portfolio and prepare its people for a new com-petitive environment. PerkinElmer’s story, though stillunfolding, illustrates the inevitable controlled chaos thataccompanies a mad dash to the future.

Let’s go back to February 1998. You were suddenly incharge of a company that needed major fixing. How didyou spend those first few months on the job?

Racing for Growth 91

HBR029ch4 8/15/01 1:35 PM Page 91

I got out in the field. I knew I had to spend time witheach of the businesses—learning about their operations,their customers, and their people. I saw a good,

proud company with anumber of real strengthsthat could be built on.We had footholds insome high-growth mar-kets such as life sciencesand digital imaging. We

had strong financial controls and a healthy balancesheet. We had good relationships with our employeesand a strong reputation with our customers. We had rea-sonably high-quality products.

But there were problems. We had an inconsistentoperating performance, a weak reputation withinvestors, a highly fragmented organization, and manybusinesses with uncertain prospects. In essence, EG&Ghad become a holding company for 31 diverse businesseswith 31 different cultures and brands. There was a lot ofgood technology in the businesses, but there was nostrategic coherence. When you put the parts together,you didn’t get a more valuable whole.

We also had a people problem. The company’s topexecutives were experienced, but some of them justdidn’t have the right skills. Many of the executives hadcome up through the government services business,where the pace and priorities were simply incompatiblewith the direction we wanted to go in. Other managers’skills were underdeveloped or too narrow as a result ofthe managers having worked in one role in one EG&Gbusiness for a long time. So while the company hadmade an intellectual commitment to move out of the

EG&G had become a holdingcompany for 31 diversebusinesses. When you put theparts together, you didn’t get a more valuable whole.

92 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 92

net30

government contracting business and into fast-pacedcommercial markets, its people were having a hard timemaking the leap.

With that kind of dire situation, where did you start?

We had to earn the right to grow. EG&G had a long trackrecord of erratic financial performance, and as a result,we lacked credibility among investors. Initially, our focuswas on improving our ability to deliver strong and con-sistent earnings. We knew that if we couldn’t gain thetrust of the financial markets, we wouldn’t be able to getthe resources—and the tolerance from investors—weneeded in order to shift our portfolio into higher-growthbusinesses. (For more on EG&G’s name change, see “AndNow for Something Completely Different” at the end ofthis article.)

The first task, then, was to establish a new culture,one with more ambitious performance goals, specificperformance metrics and rewards, and clearer account-ability. We also made a series of hard decisions in orderto create a more efficient operation. We consolidated our31 businesses into five strategic business units (SBUs),integrated sales forces, shifted production to the FarEast, developed a corporatewide materials-purchasingprogram, and raised our goals for profit and growth. Weput in place four rigorous corporatewide processes thathad been used effectively at GE and AlliedSignal: strate-gic goal setting to drive strategy; a leadership and organi-zational review to develop talent; an annual operatingplan to set performance goals and commitments; and aprocurement, quality, and productivity program for con-tinuous improvement. We offset the onetime charges


HBR029ch4 8/15/01 1:35 PM Page 93

involved in this restructuring through onetime gainsfrom the sale of several businesses, including twomechanical businesses, Sealol and Rotron. The resultslook good. Our operating margin, which was less than 6%in 1997, now exceeds 11%. Even more important, wehave consistently met or beaten our financial commit-ments for the past 12 quarters.

Once we had our operations in good shape, we movedto the next phase of transformation, which was remakingour portfolio. That involved narrowing our strategicfocus through divesting ourselves of underperformingand nonstrategic businesses. It also involved acquiringattractive new businesses that built critical mass in ourhighest-growth markets.

We used two criteria to evaluate our existing busi-nesses. First was market leadership: could the businessbecome one of the top three players in its industry ormarket? Second was growth potential: could the busi-ness produce double-digit revenue growth on an annualbasis? If the answer to either question was no, the busi-ness became a candidate for divestiture. As we wentthrough this exercise, it became clear that four of the fivebusiness units had the potential to meet our criteria: lifesciences, instruments, optoelectronics, and fluid sci-ences. The fifth SBU, government services, was profitableand well managed, and historically it had been EG&G’score strength. But it operated in a low-margin, consoli-dating, declining market. Clearly, it didn’t match the newcriteria, so in August 1999, we sold all of the governmentservices business. This was a difficult and emotionaldecision—it was like giving up half of the family—but werecognized that we wouldn’t be able to pursue the othergoals we had established unless we were willing to let goof that aspect of EG&G’s heritage.

94 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 94

Do you continue to reexamine the mix of businesseswithin each of the remaining SBUs?

Absolutely. We continue to apply the same screening cri-teria to each of the major product lines—upgrading themix by “weeding and seeding” the garden in each SBU.The healthier your portfolio, the more selective you haveto be about what’s in it.

For example, in our optoelectronics business, westarted with nine different businesses and brands butquickly narrowed our focus to just three segments: spe-cialty illumination, digital imaging, and telecommunica-tions. We accomplished this by consolidating productlines; selling two lower-growth specialty semiconductorbusinesses, Judson and IC Sensors; and acquiring LumenTechnologies, a maker of specialty illumination prod-ucts. We are now one of the market leaders in both thespecialty illumination and imaging segments, whichtogether account for 80% of the division’s revenues. Inour tele-communications segment, while we’re not yetone of the top three in the market, we’ve been able toachieve an annual revenue growth rate of more than100%.

Acquisitions such as the recent purchase of NEN LifeSciences and PE’s Analytical Instruments Division havebeen fundamental to your attempt to build criticalmass within the most attractive market segments. Howwould you describe your acquisitions philosophy?

Acquisitions are designed to quickly provide us withcomplementary products, technologies, and geographiccoverage. We have made seven key acquisitions since1998: one in optoelectronics, one in fluid sciences, two ininstruments, and three in life sciences.


HBR029ch4 8/15/01 1:35 PM Page 95

In optoelectronics, the acquisition of Lumen Tech-nologies provided two key benefits. It contributed a setof complementary specialty illumination products,which are used, for instance, in endoscopy and semicon-ductor lithography. It also allowed us to expand ouroperating margins, because we were able to shift themajority of manufacturing operations from the UnitedStates to lower-cost facilities in China, Singapore,Indonesia, and the Philippines. In fluid sciences, weacquired Belfab, a maker of seals for ultrahigh-vacuumsemiconductor equipment, which extended our reachbeyond the aerospace industry into a higher-growthindustry with a different market cycle. Yet the technol-ogy and manufacturing operations in the two sectors arevery similar. The acquisition of PE Corporation’s Analyti-cal Instruments gave us a premier brand name, a cus-tomer franchise, and a global sales and distribution net-work that covers 100 countries. Six months later, weacquired Vivid Technologies, a maker of advanced explo-sives detection systems. Together, these acquisitionshave made us number one in the world in explosivesdetection systems, such as those used behind the scenesin airports around the world, and one of the top three inanalytical instruments.

The situation in life sciences was different. There wewanted to build market share quickly in two attractive

areas. We were already aleader in genetic-diseasescreening (neonatal andprenatal testing), and wehad a good position inresearch tools for discov-

ering new drugs. We acquired Isolab, a maker of geneticscreening reagents; Life Science Resources, a maker of

Divestitures are just asimportant as acquisitions. I’m a big believer in an “invest or divest” mentality.

96 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 96

drug discovery instruments; and just recently NEN LifeSciences, a leader in reagents and consumables for drugdiscovery. We’re now the world’s third-largest providerof drug discovery tools and the leader in neonatal andprenatal testing, with reagents and consumables nowgenerating 75% of the SBU’s revenues. At the same time,we have de-emphasized our position in clinical diagnos-tics. With recent reductions in research and develop-ment spending, it now accounts for about 10% of rev-enues, down from 40% a few years ago.

We also look for acquisition synergy across the busi-nesses, such as in integrating sales networks, technology,or materials purchasing across SBUs. We also see oppor-tunities to further develop our leadership talent by pro-viding cross-functional or cross-business assignments.Finally, we look for opportunities to balance fundingcycles, so that when one business experiences a marketdownturn or goes into heavy investment mode, we canlook to other segments to counteract those effects.

But divestitures are just as important as acquisitions.I’m a big believer in an “invest or divest” mentality. Keep-ing low-growth product lines in a holding pattern dimin-ishes their potential and drags down your revenue-growth rates. The major portfolio challenge in theseacquisitions and divestitures is, how do you divest your-self of low price-to-earning businesses and acquire highP/E businesses at the same time without diluting yourearnings per share? We have been able to offset thedivestiture dilution by increasing productivity actions inour base businesses. On the acquisitions side, all of ourdeals until NEN have been accretive in the first year. TheNEN acquisition is dilutive on a GAAP basis but accre-tive on a cash EPS basis. In very high-growth markets,like drug discovery tools and fiber-optic components, the


HBR029ch4 8/15/01 1:35 PM Page 97

significant P/E premium makes GAAP dilution almostunavoidable. We probably couldn’t have done the NENacquisition a year ago without our stock price being pun-ished. But at this stage, our investors understand andsupport our strategy to upgrade the portfolio.

You talk a great deal about management talent as avaluable resource—about the need to provide chal-lenges for highly talented managers. How did you getyour people to lead change rather than resist it?

The fact is, you have to let go of those who can’t or won’tchange, bring in new people to help accelerate change,and—this is the toughest part—convert and improve thecore of the organization: the people who want to drivechange but may not have all the necessary skills.

We knew recruiting talent for the senior ranks wouldbe a challenge, given EG&G’s steady-as-she-goes reputa-tion. That’s why we had to create jobs that were highlyleveraged, both in their responsibilities and in their com-pensation packages. We also shifted power away fromthe corporate center to each of the business units. Thecorporate center became responsible for only a limitednumber of shared services such as benefits administra-tion, public company functions, and coordination of thefour corporatewide processes. That allowed us to reduceour corporate staff from 140 people in 1998 to 65 peopletoday. In the units, it allowed us to create jobs that werebold enough to attract the competitive and experiencedexecutives we needed. On the compensation side, webacked up our desire for fast-paced growth with a newperformance-based stock program for corporate officers.Vesting is based on three years of consecutive EPSgrowth of 15% or more, but if the EPS growth is 50% orgreater, the stock vests in two years.

98 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 98

Basically, we’ve created a new organization, startingwith the senior management group. We’ve reduced thenumber of officers from 15 to ten; nine are new to theposition, and seven are from the outside. If you look atour top 100 employees, you’ll see that 80% are new totheir positions, and half of those employees are new toPerkinElmer.

How have you managed to bring along existing employ-ees—the “core,” as you put it?

We try to make sure all employees have challenginggoals, support for professional development, and recog-nition for their accomplishments. And we’ve made threemajor changes to our compensation plan. First, wemoved away from an economic-value-added (EVA)model to one that is tied to profit and cash targets. Thathelped focus all our employees on tangible and easily vis-ible goals, and it simplified the previous system. Second,we raised the amount of individual employee compensa-tion that is determined by performance. Third, we cre-ated an employee stock purchase plan and turned ouroptions plan upside down. In the past, some 75% of theoptions were distributed at the corporate level, and therest were given out at the business-unit level. The reverseis true today.

We have also put into place a broad range of trainingprograms that touch all of PerkinElmer’s 13,000 employ-ees. The activities are designed to teach leadership skills,business fundamentals, and best practices. All of oursenior managers, including me, play a role in these pro-grams—either as members of the “faculty” or in give-and-take sessions in which we discuss our progress anddisappointments. (See “From Complacency to Compe-tency” at the end of this article.)


HBR029ch4 8/15/01 1:35 PM Page 99

Hundreds of books and articles have been writtenabout how to lead change and create a sense ofurgency. How do you keep the change message alive foryour employees?

Our leadership team reaches out, constantly and directly,to people throughout the organization. I spend a lot oftime at “skip level” and town-hall-style meetings withpeople in each of the businesses. Skip-level meetingsbring together a cross section of 15 to 20 employees totalk about what’s going well and what we shouldimprove. In an hour or so, I not only develop a bettersense of the business, I also have a chance to communi-cate the values and goals of the company.

I’m also a coach. It’s my job to increase the capabili-ties of leaders throughout the company. As important asrecruiting remains, re-recruiting the right people—pay-ing attention to developing people from the inside—is

essential. One way weget a picture of thestate of our people isthrough semiannualleadership and organi-zational reviews ofeach unit. Each meet-ing is attended by the

unit president and senior managers. We spend a greatdeal of time evaluating the performance of each of thetop 400 people in the company. We come away with a setof priorities about how to improve the business, whichmight include a list of employees targeted for promo-tions or developmental moves or a list of targeted newrecruits. My personal involvement in the performancereviews of the top 400 people sends a message to every-one that people are critical to PerkinElmer’s future.

Strategy and operationaldecisions are important, but the defining bets are always made on people. If you have the right person leading the charge, good things happen.

100 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 100

Looking back at your first two and a half years atPerkinElmer, were there any surprises?

It’s not so much a surprise as it is a confirmation of somegreat advice I’d been given. Seven years ago, LarryBossidy [former CEO of AlliedSignal] told me that themost important and memorable decisions are alwaysabout people. When I consider the dramatic changesPerkinElmer has already undergone and the group ofpeople who helped make those changes happen, I ameven more impressed by the truth of his advice. Strategyand operational decisions are important, but the defin-ing bets are always made on people. If you have the rightperson leading the charge, whether it’s in a factory or aresearch lab, good things happen. We earned the right togrow, and that’s why I see even more opportunities nowthan when I started this journey.

And Now for Something Completely Different . . .

Selling businesses, improving operations, and acquiringhigh-growth companies are difficult maneuvers, butattracting the attention of the right outside constituenciesis no small task either. Beginning with the company’s dra-matic name change, Greg Summe describes the chal-lenge of reintroducing PerkinElmer to Wall Street and therest of the world.

WE DECIDED TO CHANGE our name because wewanted to signal to our customers, investors, and employ-ees that this was a new company. We had alreadymade significant operational, organizational, and portfo-


HBR029ch4 8/15/01 1:35 PM Page 101

lio changes, but EG&G’s legacy kept dragging us backto a different era—at least in the minds of Wall Street andeven potential employees. We considered a number ofpossibilities, including taking on an entirely new name.Finally, in 1999, we settled on the PerkinElmer name,which we had acquired with the instruments business ear-lier that year, because it already had a prominent prod-uct brand with global recognition. We modernized thePerkinElmer logo and adopted the tag line “precisely” tocommunicate the quality of our products and the culturewe wanted to embody. At the time, the name changewas controversial. Many people were emotionallyattached to the old name, and the research we haddone on the impact of a new name was inconclusive. Buttoday the decision seems like a no-brainer because thereception has been so positive.

The name change was just part of our journey toattracting a new group of investors. Historically, EG&G’sstock was held by value-oriented investors, some ofwhom had owned EG&G stock since it had gone publicon the New York Stock Exchange in 1965. Many ofthem focused chiefly on gains in operating profit andnominal increases in the price/earnings multiple andweren’t eager to take the risks associated with portfoliochange.

As our earnings and revenue growth acceleratedand our stock price rose, our P/E multiple expandedbeyond the guidelines typically used by value investors.Therefore, they began to sell their EG&G stock, so westarted reaching out to a new community of Wall Streetanalysts and investors. Traditionally, the analysts whoresearched and reported on EG&G specialized in indus-trial, multiproduct companies. But today our research

102 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 102

coverage increasingly comes from analysts who special-ize in high-growth markets such as life sciences andtelecommunications.

The transition has enabled us to reach a broaderbase of growth-oriented investors, who support thehigher P/E multiple needed to grow our most attractivebusinesses. Today, 80% of our top 20 investors aregrowth oriented, compared to 15% early last year.

From Complacency to Competency

PERKINELMER’S TRAINING INITIATIVES, designed tobe driven by the developmental needs of individuals,consist of four programs aimed at employees at variouslevels of the organization.

Advanced Leadership Institute. Designed for middle tosenior managers, this three-day program centers on lead-ership development and business acumen. Through cus-tomized computer simulations, participants competeagainst one another in running PerkinElmer businesses.

Emerging Leaders Program. This three-day programcovers business simulations, team projects, best practices,and presentations led by senior managers. It’s aimed atearly-career managers who have been identified as hav-ing high growth potential.

Driving World-Class Performance Program. This two-day course expands on PerkinElmer’s core values andprocesses and teaches the Six Sigma Levels of Perfor-mance program, which strives for zero defects inproducts, services, and transactions. All PerkinElmer


HBR029ch4 8/15/01 1:35 PM Page 103

employees worldwide will take part in this course; thecompany’s goal is to have every employee trained in SixSigma by the end of 2001.

Skills-Based Training Programs. These programs areoffered on a continuing basis to individual employees.They include training in coaching, team building, projectmanagement, and assessment skills.

Originally published in November–December 2000Reprint R00607

104 Heimbouch

HBR029ch4 8/15/01 1:35 PM Page 104

The Tough Work of TurningAround a Team

Executive Summary

BILL PARCELLS, ONE OF the NFL’s winningest coaches,offers business leaders three rules for reversing the for-tunes of a losing team. He contends that the keys to moti-vating people are much the same whether they’re play-ing on a football field or working in an office.

The first rule is to make it clear from day one thatyou’re in charge. Parcells has found that holding frankone-on-one conversations with every member of theorganization is essential to success. Leaders can doeverything right with their teams and still fail if they aren’table to reach each member as an individual.

Rule two is that confrontation is healthy. Parcells rel-ishes confrontation because it provides an opportunity toget things straight with people. Confrontation does notmean putting someone down. When criticizing members

105

HBR029ch5 8/15/01 1:36 PM Page 105

of the team, he puts it in a positive context. One he setsthat context, he’s not afraid to be blunt about players’failings.

Parcells’s third rule is to identify small goals and hitthem. He believes that success breeds success. Once ateam gets in the habit of losing, confidence dips and suc-cess seems unreachable. To break the habit of losing,Parcells focuses on achieving goals within immediatereach.

In the end, Parcells is convinced that if you get peo-ple on your team who share the same goals and thesame passion, and if you push them to achieve at thehighest level, you’re going to come out on top.

T have little loyalty;some even want you to fail. Your star performers expectconstant pampering. Your stockholders are impatient,demanding quick results. And the media scrutinize andsecond-guess your every move.

I can relate.As a coach in the NFL, I’ve been in a lot of pressure-

cooker situations, and my guess is that the challengesI’ve faced are not all that different from the ones thatexecutives deal with every day. I’m not saying that busi-ness is like football. I am saying that people are people,and that the keys to motivating them and getting themto perform to their full potential are pretty much thesame whether they’re playing on a football field or work-ing in an office.

The toughest challenge I’ve faced as a coach is takinga team that’s performing poorly and turning it around.I’ve done it three times now. In 1983, my first year as a

106 Parcells

HBR029ch5 8/15/01 1:36 PM Page 106

head coach, I led the New York Giants through anabysmal season—we won only three games. In the nextsix seasons, we climbed to the top of the league, winning

two Super Bowls. When Ibecame coach of the NewEngland Patriots in 1993,they were coming off twoyears in which they’d won acombined total of threegames. In 1996, we were inthe Super Bowl. In 1997,when I came to the New

York Jets, the team had just suffered through a 1-15season. Two years later, we made it to the conferencechampionship.

Those turnarounds taught me a fundamental lessonabout leadership: You have to be honest with people—brutally honest. You have to tell them the truth abouttheir performance, you have to tell it to them face-to-face, and you have to tell it to them over and over again.Sometimes the truth will be painful, and sometimes say-ing it will lead to an uncomfortable confrontation. So beit. The only way to change people is to tell them in theclearest possible terms what they’re doing wrong. And ifthey don’t want to listen, they don’t belong on the team.

Taking Charge

To lead, you’ve got to be a leader. That may sound obvi-ous, but it took me an entire year to learn—and it wasn’ta pleasant year. When I started as coach of the Giants, Ilacked confidence. I was surrounded by star players withbig names and big egos, and I was a little tentative indealing with them. I didn’t confront them about how

The only way to changepeople is to tell them in theclearest possible termswhat they’re doing wrong.And if they don’t want tolisten, they don’t belong onthe team.


HBR029ch5 8/15/01 1:36 PM Page 107

net30

they needed to change to succeed. As a result, I didn’t gettheir respect and I wasn’t able to change their attitudes.So they just kept on with their habit of losing.

At the end of the season, I figured I’d be fired. Butmanagement ended up asking me back for another sea-son—mainly because they couldn’t find anybody toreplace me. At that point, I knew I had nothing to lose, soI decided I would do it my way. I was going to lead andthe players were going to follow, and that’s all there wasto it. On the first day of training camp, I laid it on theline: I told everyone that losing would no longer be toler-ated. Players who were contributing to the team’s weakperformance would be given a chance to change, and ifthey didn’t change, they’d be gone.

It was a tough message, but I balanced it with a morepositive one. I told them what I think a team is all about:achievement. Sure, they could make a lot of money infootball and they could buy a lot of nice things, but theonly permanent value of work lies in achievement, andthat comes only with relentless effort and commitment.It wasn’t going to be easy, but at the end of the day,achievement would be the most important thing theywould take home with them.

After I talked to them as a group and established mycredibility as a leader, I began talking with them person-ally. With the Giants, and with the other teams I’vecoached, I’ve found that holding frank, one-on-one con-versations with every member of the organization isessential to success. It allows me to ask each player forhis support in helping the team achieve its goals, and itallows me to explain exactly what I expect from him. I tryto appeal to the players’ passion for achievement andwinning, but I’m also very clear that if they don’t give the

108 Parcells

HBR029ch5 8/15/01 1:36 PM Page 108

team what it needs, then I’m going to find someone elsewho will. I tell them, “If you don’t want to play in thechampionship games and you don’t want to achieve atthe highest level, then I don’t want you here, becausethat’s what I’m trying to do. I am not trying to finishfourth.” Leaders can do everything right with their teamsand still fail if they don’t deliver their message to eachmember as an individual.

Those conversations also give me a basis for makingan honest evaluation of every player. It’s all too easy tocome into an organization that’s been struggling andmake blanket judgments about everybody—to thinkeverybody’s failing. But that’s a mistake. There can bemany hidden strengths on a team, just as there can bemany hidden weaknesses. The only way you can bringthem to the surface is by watching and talking with eachteam member. You’ll quickly see who’s a contributor andwho’s an obstacle. And, for the good of the team, you’llwant to move swiftly to get the obstacles out of the way.The hard fact is that some people will never change.

So if you’re called in to turn around a team, here’sRule One: make it clear from day one that you’re incharge. Don’t wait to earn your leadership; impose it.

The Power of Confrontation

If you want to get the most out of people, you have toapply pressure—that’s the only thing that any of us reallyresponds to. As a coach, I’ve always tried to turn up theheat under my people, to constantly push them to per-form at a high level.

Creating pressure in an organization requires con-frontation, and it can get very intense, very emotional.


HBR029ch5 8/15/01 1:36 PM Page 109

I’ve seen coaches avoid confrontations with their play-ers because they don’t like conflict, and I assume thesame thing is true among the leaders of business teams.But I’ve actually come to relish confrontation, not be-cause it makes me feel powerful but because it providesan opportunity to get things straight with people. It’snot until you look people right in the eye that you getto the sources of their behavior and motivation. With-out confrontation, you’re not going to change the waythey think and act.

Confrontation does not mean putting someone down.When you criticize members of the team, you need toput it in a positive context. I’ve often said to a player, “I

don’t think you’re per-forming up to your poten-tial; you can do better.”But I also made it clearthat my goals were hisgoals: “It’s in your bestinterest that you succeed,

and it’s in my best interest that you succeed. We reallywant the same thing.” Once you set that context, though,you shouldn’t be afraid to be blunt about people’s fail-ings. You shouldn’t be afraid to offend them. You need todo what it takes to get a strong reaction because thenyou know you’ve reached them.

In the end, I’ve found, people like the direct approach.It’s much more valuable to them to have a leader who’sabsolutely clear and open than to have one who soft-soaps or talks in circles. I’ve had many players come backto me ten years later and thank me for putting the pres-sure on them. They say what they remember most aboutme is one line: “I think you’re better than you think youare.” In fact, they say they use the same line with their

Even as I’m confronting players about their weaknesses, I’m also alwaystrying to rebuild a culture of success.

110 Parcells

HBR029ch5 8/15/01 1:36 PM Page 110

kids when they’re not doing so well in school or are hav-ing other problems. My father used that expression withme, and there’s a lot of truth to it—people can do morethan they think they can.

That’s Rule Two: confrontation is healthy.

Success Breeds Success

The prospect of going from a team that’s at the bottom of the standings to one that’s on top is daunting. Whenyou’ve done a lot of losing, it gets hard to imagine yourselfwinning. So even as I’m confronting players about theirweaknesses, I’m also always trying to build a culture ofsuccess. That’s not something you can do overnight. Youhave to go one step at a time, the same way you move theball down the field, yard by yard.

Here’s my philosophy: to win games, you need tobelieve as a team that you have the ability to win games.That is, confidence is born only of demonstrated ability.This may sound like a catch-22, but it’s important toremember that even small successes can be extremelypowerful in helping people believe in themselves.

In training camp, therefore, we don’t focus on the ulti-mate goal—getting to the Super Bowl. We establish aclear set of goals that are within immediate reach: we’re going to be a smart team; we’re going to be a well-conditioned team; we’re going to be a team that playshard; we’re going to be a team that has pride; we’re goingto be a team that wants to win collectively; we’re going tobe a team that doesn’t criticize one another.

When we start acting in ways that fulfill these goals, Imake sure everybody knows it. I accentuate the positiveat every possible opportunity, and at the same time Iemphasize the next goal that we need to fulfill. If we have


HBR029ch5 8/15/01 1:36 PM Page 111

a particularly good practice, then I call the team togetherand say, “We got something done today; we executed realwell. I’m very pleased with your work. But here’s what Iwant to do tomorrow: I want to see flawless specialteams work. If you accomplish that, then we’ll be readyfor the game on Sunday.”

When you set small, visible goals, and people achievethem, they start to get it into their heads that they cansucceed. They break the habit of losing and begin to getinto the habit of winning. It’s extremely satisfying to seethat kind of shift take place in the way a team thinksabout itself.

So Rule Three is: set small goals and hit them.

Picking the Right People

Another challenge in building a winning team comesfrom free agency. I know that companies today are hav-ing trouble hanging on to their best people; there’s agreat deal of turnover and not much loyalty. That’s a sit-uation that I had to adapt to as a coach.

One of the things that initially helped me become suc-cessful in the NFL was my ability to develop players withthe Giants. We had a program in place, and we broughtpeople along slowly. Today, you no longer have the time todevelop your talent in the old way. The situation is morelike coaching high school football in some respects—every year, the senior class graduates and moves on.When I started, coaches reworked maybe 8% or 10% oftheir teams every year. Now it’s sometimes as high as 30%.

That kind of turnover adds a tough new wrinkle toturning a team around and keeping it on the winningtrack. In particular, you have to be extremely carefulabout the new people you bring on. You can do seriousdamage with a few bad choices. Unfortunately, there’s no

112 Parcells

HBR029ch5 8/15/01 1:36 PM Page 112

science to picking the right people. There’s a lot of trialand error involved. You’re going to get fooled by people,and you’re going to make mistakes—I know I’ve mademy share. But after a while, you start to develop a senseof who’s likely to work out. I’ve found it’s not always theone who has the best reputation or even the most out-standing set of talents. It’s usually the one who under-stands what it will take to succeed and is committed tomaking the effort.

For example, there’s a player, Bryan Cox, who had aterrible reputation in the NFL. He’d been fined a lot ofmoney by the league—maybe more than anyone in itshistory. My teams had played against him so many timesthat I almost felt like I knew him. And watching himplay, I’d say to myself, this guy plays so hard and tries sohard—he’s got something that I want to have on myteam. So when he was a free agent, I called him on thephone and we had a straight, tough talk. I told himexactly what I wanted from him, and he told me what hewanted from me, which boiled down to this: “Don’t BSme.” I told him he’d always know what I was thinking.Bryan signed on with the Jets, and he’s done a great jobfor the franchise.

I’m no psychologist. I don’t care about what kind ofpersonality someone has or whether it corresponds withmy own. I don’t care if they’re “well adjusted.” I just wantmy players to want to win as much as I want to win. I’mconvinced that if you get people onto your team whoshare the same goals and the same passion, and if youpush them to achieve at the highest level, you’re going tocome out on top.



HBR029ch5 8/15/01 1:36 PM Page 113

Saving Money, Saving Lives

Executive Summary

IN 1996, DUKE CHILDREN’S HOSPITAL was in serioustrouble. Its $11 million annual operating loss had forcedadministrators to make cutbacks. As a result, some care-givers felt that the quality of care had deteriorated. Par-ents’ complaints were on the rise. Frustrated staff mem-bers were quitting.

In this article, Jon Meliones, DCH’s chief medicaldirector, candidly describes how his debt-ridden hospitaltransformed itself into a vibrant and profitable one. Theproblem, he realized, was that each group in DCH wasfocusing only on its individual mission. Doctors and nurseswanted to restore their patients to health; they didn’t wantto have to think about costs. Hospital administrators, fortheir part, were focused only on controlling widely esca-lating health care costs. To keep DCH afloat, cliniciansand administrators needed to work together.

115

HBR029ch6 8/15/01 1:37 PM Page 115

By listening to staff concerns, turning reams of confus-ing data into useful information, taking a fresh approachto teamwork, and using the balanced scorecard method,Meliones and his colleagues brought DCH back to life.Developing and implementing the balanced scorecardapproach wasn’t easy: it took a pilot project, a top-downreorganization, development of a customized informa-tion system, and systematic work redesign. But theirefforts paid off. Customer satisfaction ratings jumped18%. Improvements to internal business processesreduced the average length of stay 21% while the read-mission rate fell from 7% to 3%. The cost per patientdropped nearly $5,000. And DCH recorded profits of$4 million in 2000.

This first-person account is required reading for anyexecutive seeking to revitalize a sagging organization.Meliones shares the operating principles DCH followedto become a thriving business.

M seven o’clock on a hecticNovember evening in 1996. I was the attending physicianin the intensive care unit at Duke Children’s Hospital(DCH) in Durham, North Carolina. A six-month-oldnamed Alex lay in a crib in the ICU with a stiff plastictube in her throat. Awake and moving after heartsurgery, the tiny girl was ready to come off the ventilator.As Alex squirmed and tried to breathe, the ventilatorforced more air into her lungs. Her exhausted parentsgrew distraught. “Why can’t she come off the ventilator?”her mother asked. “Because we’ve had to cut back onnight staff,” replied the busy nurse. “There’s no respira-tory therapist available.” Alex was uncomfortable. Shereceived medication to help her sleep and to keep her

116 Meliones

HBR029ch6 8/15/01 1:37 PM Page 116

from fighting the ventilator until the therapist arrived inthe morning. But her parents didn’t sleep; they were tooconfused and upset.

As I watched Alex and her parents, I thought back tosimilar scenes I had witnessed over the years at DCH, a134-bed pediatric hospital located on the fifth floor ofDuke University Hospital. Here, 800 employees care forpatients in our neonatal intensive care unit, pediatricintensive care unit and pediatric emergency room, bone-marrow transplant and intermediate care units, as wellas in our subspecialty and outreach clinics. When I cameto DCH in 1992, we had a $4 million annual operatingloss; it had grown to $11 million by 1996, which forcedadministrators to cut back on resources. As a result,some caregivers felt that the quality of clinical care haddeteriorated. Parents’ complaints increased. Some dis-satisfied doctors threatened to send their patients else-where. Frustrated staff quit.

And then it struck me. I saw with perfect clarity thereason that DCH was struggling to meet the needs of itscustomers—our patients and their parents. And I knewwhat had to be done to make things right. The problemwas that our hospital was a collection of fiefdoms: eachgroup, from accountants to administrators to clinicians,was focusing on its individual goal rather than on theorganization as a whole. We would be a far more effec-tive organization if we could stop that from happening.Most companies in the United States had this insight 20years ago, but the nonprofit world remains, for the mostpart, unaware of it. I realized that DCH needed to startthinking less like a money-losing nonprofit and more likea profitable corporation.

A sense of mission, of course, is critical to any organi-zation’s identity. The institutional mission of a hospitalis to promote the health of the community. But during


HBR029ch6 8/15/01 1:37 PM Page 117

difficult periods, it’s easy to lose sight of the big pictureand focus solely on your fiefdom’s specific goals. Clini-cians—that is, doctors and nurses—want to restore theirpatients to health; they don’t want to think about costs.Hospital administrators have their own mission—tocontrol wildly escalating health care costs.

Cost cutting in a vacuum traumatizes patients, frus-trates clinicians, and ultimately cripples the hospital’smission. The decision to cut a respiratory therapist fromthe night shift, for example, affected Alex and her par-ents as well as their insurance company, which had topay an additional $2,000 to cover the cost of the ventila-tor and ICU care. The decision also left the cliniciansfeeling powerless, since decisions regarding clinical prac-tice were being made without their input. Such trade-offsbetween quality of patient care and cost control causeintense conflict for health care professionals. In worst-case situations, efforts to improve profit margins actuallyhave the opposite effect—they chase away customers,cost executives their jobs, and put the entire hospital atrisk of financial ruin.

Regaining Our Balance

Considering the magnitude of the issues we faced—a $7million increase in annual losses in four years—it’s hardto believe that we ever turned things around. But we did,by changing people’s minds and hearts, inch by inch, dayby day. In 1997, the chief nurse executive, nurse man-agers, and I began working together to start turning theorganization around. First, we discussed our currentrealities with the entire clinical team. We opened themeetings by talking about our goals for our patients. “Wewant patients to be happy,” the doctors and nurses

118 Meliones

HBR029ch6 8/15/01 1:37 PM Page 118

agreed, “and for them to have the best care.” We alsodescribed our pressing financial challenges.

We showed the clinicians our raw data. The averagelength of stay at DCH was eight days—20% longer thanthe six-day national average. The average per-patientcost was $15,000—more money than we were bringingin. If we continued to spend at the same rates, wewould be forced to cut clinical programs, staff, andbeds. The quality of patient care and our reputationwould then suffer, and we would fail to meet the needsof our community.

Confronted with this grim picture, the cliniciansbegan to understand that if we wanted to save our pro-grams and our patients, create an environment in whichstaff are fulfilled, and keep our jobs, we would all have toreadjust our individual missions and start paying atten-tion to costs. If the hospital didn’t show a margin, clini-cians wouldn’t be able to fulfill their mission. Thus, weadopted the now-familiar mantra in health care: no mar-gin, no mission.

It was also clear that the administrators needed to behighly involved. To bring the administrators’ and theclinicians’ missions into alignment, we turned to a prac-tical management approach that had worked well innumerous Fortune 500 corporations: the balanced score-card method. Developed by Robert Kaplan and DavidNorton, it had improved customer service, driven organi-zational change, and boosted bottom-line performancein leading companies like AT&T, Intel, and 3M. Our goalwas to become the health care leader in the balancedscorecard. (See “A Look at the Numbers” for DCH’sresults.)

Our balanced scorecard aligned the hospital’s goalsalong four equally important quadrants: financial health;


HBR029ch6 8/15/01 1:37 PM Page 119

Net margin

-$11

-$8

-$6

1997 1998 1999 2000

$4

$14,889

$13,411

$12,550$12,440

$10,500

20001999199819971996

Cost per case

A Look at the Numbers

Using the balanced scorecard method, Duke Children’s Hospital’s cost-per-case average fell from nearly $15,000 to $10,500 and its marginsoared from an $11 million annual loss to a $4 million gain.

HBR029ch6 8/15/01 1:37 PM Page 120

customer satisfaction; internal business procedures; andemployee satisfaction. We explained the theory to clini-cians and administrators like this: if you sacrifice toomuch in one quadrant to satisfy another, your organiza-tion as a whole is thrown out of balance. We could, forexample, cut costs to improve the financial quadrant byfiring half the staff, but that would hurt quality of service,and the customer quadrant would fall out of balance. Orwe could increase productivity in the internal businessquadrant by assigning more patients to a nurse, butdoing so would raise the likelihood of errors—an unac-ceptable trade-off. Our vision, which became the newmission statement, was to provide patients and familieswith high quality, compassionate care within an efficientorganization.

Taking Our Medicine

Developing and implementing a balanced scorecard islabor intensive because it is a consensus-driven method-ology. To make ours work required nothing short of apilot project, a top-down reorganization, development ofa customized information system, and systematic workredesign. The most difficult challenge was convincingemployees that they must work in different ways.

At first, doctors and managers saw attempts to movethem into teams as a shift in their power base. Nearlyeveryone complained that applying a systematicapproach to cost management was “cookbookmedicine.” It took a good deal of persuasion, persistence,and reassurance to get some individuals to buy into ourprocess. One cardiologist routinely stormed out of meet-ings when we talked about cost per case.

We knew that changing people’s minds would be hardwork. But once people saw how successful the balanced


HBR029ch6 8/15/01 1:37 PM Page 121

scorecard approach was in one area of the hospital, wereasoned, it would be easier to sell the methodologythroughout the rest of the organization. So we decided tolaunch a pilot project. Some physicians were much morewilling to change than others. Those who understood theimportance of applying systems to medicine—such assurgeons—became our first champions. So we startedthe balanced scorecard in one very important microcosmof the hospital—the pediatric intensive care unit, whichI lead.

First, we reorganized the roles that individuals play inthe ICU. We moved from mission-bound departments inwhich people identified only with their particular jobs (“I am a manager,” “I am a nurse,” and so on) to goal-oriented, multidisciplinary teams focused on a particularillness or disease (“We, the ICU team, consisting of themanager, the nurse, the physician, the pharmacist, andthe radiologist, help children with heart problems”). Wecalled these teams clinical business units—what otherindustries call business or operating units. The leadphysician and the lead administrator shared responsibil-ity in these teams. Together, they reviewed financialinformation, patient and staff satisfaction data, andinformation on health care trends and initiatives.

The various clinical business units worked together toorganize “care coordination rounds” and brainstormsolutions to difficult patient cases. They created apatient’s care plan—a document, shared with the par-ents, that records everything from treatment recommen-dations to post-hospital care.

The teams also developed protocols we call clinicalpathways—a set of best practices for various treatments.For example, a respiratory therapist, a nurse, and aphysician developed a series of steps a nurse could follow

122 Meliones

HBR029ch6 8/15/01 1:37 PM Page 122

to remove a patient from a respirator without having atherapist present. As the clinicians developed new path-ways, they shared their successes with the entire organi-zation so we could all learn from their experience.

By developing and promoting protocols like these, weimproved care dramatically. For example, we knew thatbabies recovering from heart surgery had trouble feeding

and that parents needed tolearn how to help them.Before we had formed thepathways, we would waituntil the day of discharge toteach parents how to do so.Once people started shar-ing their expertise todevelop the pathways, we

learned that there was no reason to wait so long andmoved the training to the day after surgery. Patientswere able to go home much sooner, and their hospitalcosts were cut by 28%.

We developed more protocols by comparing patientdata. A study of 20 heart patients, for example, revealedthat treatment costs varied dramatically. One childreceived two days’ worth of antibiotics; another receivedseven days’ worth for the same condition. One childunderwent ten laboratory tests; another had only three,and so on. As a group, the clinicians went over each case,comparing notes and reviewing the medical literature.They decided which tests were unnecessary and elimi-nated them.

Within six months, our balanced scorecard approachin the ICU was garnering impressive results. We reducedthe cost per case by nearly 12% and improved our mea-sured patient satisfaction by 8%. In fact, our pilot project

Within six months, our balanced scorecard approach reduced the cost per case in the ICU by nearly 12% and improved our patient satisfaction by 8%.


HBR029ch6 8/15/01 1:37 PM Page 123

was working so well that we implemented it in pedi-atrics, then in all of the other areas of DCH, within a year.We didn’t use a cookie-cutter approach; rather, leadersin each unit customized the scorecard template for theirspecific areas.

Over time, even the physician who had angrily left ourinitial meetings began to find ways to lower his cost percase without compromising patient care. For example,instead of keeping some patients awaiting surgery in thehospital, he discharged them overnight to a nearby hotel,lowering the total cost by $1,000 per day while makingthe patients and their parents much more comfortable.

A Measure of Progress

Like most hospitals, DCH collects a tremendous amountof data. We rigorously detail things like length of stay,number of staff, cost per case, and so on. But we wereculling very little useful information from the data—andsome of it was false. For example, the first report card onmy own performance showed that I had discharged 70patients with an average length of stay of 29 days and anaverage cost per case of $70,000. Taken together, thesenumbers deserved a grade of F. I knew that since I’d beenhead of the intensive care unit, I’d cared for and trans-ferred 1,500 patients. What was going on here? A closerlook at the data revealed that they reported on only the70 patients who had died, not my total caseload.

Clearly, we needed to approach the data in a new wayand turn it into useful information. Unless we did, wewouldn’t know where our potential cost savings were.We didn’t know, for example, that babies were needlesslykept on $2,000 ventilators at night, nor did we know howmuch that decision was costing the hospital. So for every

124 Meliones

HBR029ch6 8/15/01 1:37 PM Page 124

clinical business unit, we created a measurement systemfor each of the four balanced scorecard quadrants.

To measure our progress, we asked our IT departmentto help us develop our own database and cost-accountingsystem. Using information pulled from national data-bases, we determined national averages for indicatorssuch as length of stay and complication rates. (In 1997,custom development was our only option. We’ve sinceinstalled StrategicVision software from SAS to supportour extensive data management, trend analysis, andperformance reporting needs.) The system logged eachpatient’s treatment history and costs for everything froma $15 hypodermic needle to a $5,000 heart-lung bypassoperation. The system also tracked the average waitingtimes for admission and discharge, blood culture contam-ination rates, and so on.

The new system helped us find ways to improve ourperformance in each of the four quadrants. Many of thesteps we took were small, but cumulatively, they made abig difference. For example, our clinical pathwaysincluded a “patient care guide” for parents thatexplained in lay terms what they could expect to happenon a daily basis during their child’s hospital stay. We alsolearned from our customer surveys that parents felt frus-trated by not knowing who their child’s attending physi-cian or nurse was at any given time. So we simply putidentification cards on the doors, naming the attendingdoctor and primary nurse. Our customer satisfactionscores rose sharply.

We made other changes, too. For the financial quad-rant, for example, we reviewed the most significant datapoints, such as the number of patients admitted, treated,and released, and the cost per patient. The clinical busi-ness units reviewed cases of patients whose diagnostic,


HBR029ch6 8/15/01 1:37 PM Page 125

surgical, pharmacy, and postoperation costs had beenthe highest, and tried to determine why. In many cases,our research showed us new ways to do business. Forexample, we learned that children often stayed longerthan necessary in our $1,700-per-day ICU, in which thenurse to patient ratio is 1 to 1 or 1 to 2. That was because

the patients weren’t quiteready to move to the regu-lar pediatric floor, wherethe ratio of nurses topatients is 1 to 5 and thecost is $700 per day. So wecreated a six-bed, $1,200-

per-day transitional care unit, where the nurse to patientratio is 1 to 3. Patients could stay there until they couldbe moved to the general floor. Not only did our cost-per-patient numbers drop but also our patients’ familiesgot to spend more time with their recovering children.

Overall, the results we’ve achieved at DCH by usingthe balanced scorecard have been stunning. By increas-ing the number of clinical pathways and communicatingmore with parents, our customer satisfaction ratingsjumped by 18%. Improvements to our internal businessprocesses reduced the average length of stay from 7.9days in 1996 to 6.1 days in fiscal year 2000, while thereadmission rate fell from 7% to 3%. And employeesnoted a 45% increase in satisfaction with children’s ser-vices and with the way the entire administrative teamperformed its job.

Impressive results occurred on the financial front,too. The cost per patient dropped by nearly $5,000—afact not lost on parents, insurers, and our own seniorleaders. By FY 2000, we had gone from $11 million in

The cost per patient dropped by nearly $5,000—a fact not lost on parents, insurers, and our own senior leaders.

126 Meliones

HBR029ch6 8/15/01 1:37 PM Page 126

losses to profits of $4 million, even though we wereadmitting more patients. We achieved a reduction incosts of $29 million over these four years, without staffcutbacks. Our methodology has proved so successfulthat the entire Duke University Hospital now uses it asa framework. With the balanced scorecard we havedrastically improved our margin and achieved our hos-pital’s mission. (For more on how to turn your organi-zation around, see “Survival Strategies” at the end ofthis article.)

Lessons Learned

Yes, DCH has navigated a tremendous turnaround, but Idon’t want to suggest that it’s been easy. Adopting the

balanced scorecardapproach presented us withhuge management chal-lenges on a daily basis. In theearly stages, we often foundit difficult to keep discus-sions on target. We spentnearly a month debatingwhether a certain goal or tar-get belonged in the internal

business process quadrant or the customer satisfactionquadrant. We learned to limit those discussions—it wastoo easy to get embroiled in semantics and lose our focuson patients and staff.

We also found that people became demoralized if wecompared their performance to an abstract or too-loftytarget. For that reason, we encouraged employees to usetheir own performance as the primary benchmark. Still,

There’s a fine art to com-municating with professionals who know more than you do about their particular subject and who are passionate about their work.


HBR029ch6 8/15/01 1:37 PM Page 127

if they wanted to see how their performance comparedwith the hospital as a whole, or with a national average,they could review those data points as well.

We learned to set our targets conservatively at first:an annual 10% reduction in the length of stay was some-thing most of us felt comfortable reaching for, but a goalof 20% would have been too intimidating. As we becamemore successful, we set more aggressive targets.

And I learned that there’s a fine art to communicatingwith professionals who know more than you do abouttheir particular subject and who are passionate abouttheir work. You can’t just order them around. You haveto get inside their heads and figure out what they’regoing through.

Before 1996, I thought I was a decent communicator.But over time, I’ve had to learn to listen carefully notonly to what people are telling me but also to what I’msaying to them. Today I know that I can’t make a point ina conversation by talking in the abstract. I have to saysomething that personally matters to the other individ-ual. I learned not to say things like, “Duke Children’sHospital is losing $11 million per year.” Rather, I openedconversations with a question, such as “How importantdo you think it is to have a therapist on this unit to workwith your patients?” When they said it was important, I’dfollow up with “How can we work together to manageour costs so we can preserve the therapist’s job?”

I learned that little things make a big difference whenit comes to morale building. We created all kinds of com-munication and feedback mechanisms. I started anewsletter, “Practicing Smarter,” so staff members couldshare best practices and keep one another apprised oftheir progress. We honored “team members of themonth,” started on-line discussion groups, and spon-

128 Meliones

HBR029ch6 8/15/01 1:37 PM Page 128

sored a series of staff brown-bag lunches and openforums. These approaches may sound simple, but theyreally did help to change our culture. For the first time,employees felt that their opinions mattered.

I discovered how important it is to share the pulpitduring dramatic organizational changes. Not only did Irespect the chief nurse executive, the managers, and theadministrators as partners, but I knew that they couldcommunicate more effectively with their own con-stituencies than I ever could.

Even in the most earnest conversations, I’ve foundthat having a sense of humor is essential. For example, Ideveloped a Letterman-style list of the “Top Ten Reasonsfor Using the Balanced Scorecard,” poking fun at myselfin meetings. Once, I even walked through the hospitaldressed up as the eminently poke-able Pillsbury Dough-boy. Keeping things light made it easier for us all toendure the tremendously challenging course we’d set forourselves.

I learned, too, to respect the persuasive power ofmeaningful information. I spent hours with members ofour IT department, telling them what the staff wastelling me—trying to slice and dice our enormous moun-tains of data into useful information. When we finallypresented people with accurate tracking measures abouttheir personal performance, they were fascinated—andanxious to improve.

It’s been four years since we set out to improve perfor-mance at Duke Children’s Hospital, and changes are stillhappening. We talk about our scorecard constantly; we’refine-tuning what works and discarding what doesn’t.Whenever a clinician comes up with a better pathway, we spread the word through our newsletter and on ourbulletin boards.


HBR029ch6 8/15/01 1:37 PM Page 129

Of all the changes that have occurred, the most tellingare the ones we see in our patients. Consider the case ofRyan, a four-month-old who recently recovered fromheart surgery. At 8 P.M., Ryan was breathing with a venti-lator—just as Alex had—and his parents kept vigil by hiscrib. But unlike Alex’s parents, Ryan’s parents knewexactly who was responsible for their child’s care, whathis care entailed, and that he’d soon be transferred to anintermediate care unit. At 9 P.M., Ryan began breathingon his own. The nurse skillfully removed the plastic tubeand gently placed him on his mother’s lap. For me, see-ing Ryan sleeping peacefully in his mother’s arms was arewarding end to a long, hard, but ultimately satisfyingjourney.

Survival Strategies

The challenges faced by Duke Children’s Hospital areby no means unique to the health care industry. Indeed,many organizations find themselves in similar situations.They fear that focusing on costs will compromise theirhigher mission of serving the community—but in fact, astrong bottom line will make fulfilling their missions thatmuch easier. If you’re trying to turn your organizationaround, you may want to adopt the operating principleswe followed to make DCH a thriving business.

Communicate, Communicate, Communicate

• If your organization is in trouble, be honest. Make itabsolutely clear to everyone in the company that survivaldepends on cost management.

130 Meliones

HBR029ch6 8/15/01 1:37 PM Page 130

• Listen to what employees are saying; they know theirjobs better than you do. Instead of issuing orders, askthem, “What can we (as an organization) do?”

• Share the pulpit. People with other expertise can helpbuild consensus.

• Change people’s roles; instead of identifying with anindividual job (“I am a nurse”), employees should identifywith goal-oriented teams (“We, the ICU team, worktogether to help children with heart problems”).

• Offer constant feedback. Frequent evaluations help keepthe organization on track.

• Publicly celebrate every employee and team success.

• Cultivate your sense of humor—people will respond if youcan laugh at yourself.

Chart Your Path

• Start with a pilot project; succeeding in one departmentwill pave the way for organizationwide change.

• Set conservative goals at first; you’ll gain the confidenceneeded to set more aggressive targets.

• Focus on a few key goals; changing everything at onceleads to failure.

• Turn data into information. Work with your informationtechnology people to ensure that employees can cor-rectly interpret measurements and statistics.

• Let employees compete with their own performance, notwith some abstract competitive or statistical target.

Never Stop

• When mapping your business to the balanced score-card, don’t get sidetracked by semantics.


HBR029ch6 8/15/01 1:37 PM Page 131

• Be willing to experiment; learn from failures.

• Constantly revise and improve practices.

• Encourage strategic thinking at all levels.


132 Meliones

HBR029ch6 8/15/01 1:37 PM Page 132

Harley’s Leadership U-Turn

Executive Summary

WHEN RICH TEERL INK, now retired chairman and CEOof Harley-Davidson, became president and COO of itsmotorcycle division in 1987, the hard work of saving thecompany was already done. The company had just sur-vived seven arduous years of crisis and was on a steadypath toward growth.

Realizing how much circumstances had changed,Teerlink immediately saw the need for a new kind of lead-ership—one far removed from the command-and-controlmodel that had carried it throughout its turnaround. Whatthe company needed instead was leadership that wouldencourage employees to break new ground—an environ-ment where both union and salaried employees wouldwant to do better, where they’d care about the companyon a personal level and work together to improve bothindividual and overall performance.

133

HBR029ch7 8/15/01 1:37 PM Page 133

In this First Person account, Teerlink tells the story ofhow difficult it was to make that shift. Even if you arehardwired (like he is) to share power rather than exert it,he says, it’s extremely hard to avoid the top-downapproach. In a plan to improve operations, for example,Teerlink came up with an idea for a program he hopedwould motivate every employee, customer, and stake-holder. The program, called gain sharing, would allowevery employee to share in the company’s financial suc-cess. The problem? As his then consultant Lee Ozleypointed out, the program was yet another initiative cre-ated and imposed from the top.

Throughout his tenure, Rich Teerlink launched severalprograms designed to elicit ideas, enthusiasm, concerns,and vision from employees. These initiatives—as well asother efforts toward inclusiveness—have helped transformHarley’s culture and made it the success story it is today.

W and COO of Harley-Davidson’s motorcycle division in 1987, the hard work ofsaving the company was done. We had survived sevenarduous years of crisis. But that hardly meant Easy Streetlay ahead. In fact, we faced an altogether new and daunt-ing challenge—sustaining progress. But how to do that?For me, the answer was clear—especially when it cameto issues of leadership. We needed to create an environ-ment at Harley where everyone took responsibility forthe company’s present and its future.

I knew that such an approach wouldn’t come natu-rally to Harley. After all, our crisis had been managedwith an unmistakable top-down approach, as is so oftenthe case with turnarounds. But now that times had

134 Teerlink

HBR029ch7 8/15/01 1:37 PM Page 134

changed, so, too, could our way of doing things. Ibelieved then, and still do, that people are an organiza-tion’s only sustainable competitive advantage. Theleader should mind the interests of all stakeholders, ofcourse, but he or she should also be an outspoken advo-cate for employees, making sure they are front and cen-ter in an organization.

It’s important for people to understand that even ifyou are hardwired, like me, to be a leader who sharespower rather than exerts it—even if you set out to be a

listener and a teamplayer—the command-and-control model ishard to avoid. That’sbecause the top manage-ment job carries certainexpectations on behalf of

employees, colleagues, and the outside world. It takestrust on the part of employees and discipline on the partof the leader to push back on those traditional expecta-tions and create a company where decisions andaccountability are owned by all.

Survival Mode

My career with Harley-Davidson dates back to August1981 when I joined as CFO. I began my new job just twomonths after a group of 13 Harley managers had boughtthe company from its then parent company, AMF, in aleveraged buyout. I knew I was walking into a tough situ-ation. Harley-Davidson was certainly an American icon,but back then it wasn’t having any success in the mar-ketplace. It had a poor reputation for quality and reliabil-ity, and it was behind the curve on product design and

Even if you are hardwired tobe a leader who shares power rather than exerts it, the command-and-control model is hard to avoid.


HBR029ch7 8/15/01 1:37 PM Page 135

development. Break-even points were high and, not sur-prisingly, market share was falling. What’s more, thecompany faced some of the toughest competitors in theworld: Honda and Yamaha. It’s no wonder that in 1980,when AMF engaged an investment bank to sell the divi-sion, it found no takers.

When an organization is under extreme pressure—somuch so that one wrong move can mean its collapse—authoritarian leadership may very well be necessary. Itcertainly allows managers to act fast. Vaughn Bealsneeded speed, and so that’s the kind of leadership stylehe tended to use. Indeed, it came to him quite naturally.

Over the next few years, Vaughn and the leadershipteam (of which I was member) cut the overall workforceby 40%—affecting both the salaried ranks and the hourlyworkforce. All remaining salaried employees took a 9%pay cut and agreed to have their pay frozen at thosereduced levels for at least two years. This change, andmany others, originated at the top of the corporate pyra-mid, with no room for push back from the ranks below.

By the mid-1980s, it appeared that Vaughn and theteam might pull off the miracle Harley needed. The com-pany had introduced its Evolution Engine in 1984, andthat engine, combined with the new Softail product line(an elegant variation on the classic Harley look), beganto make some money. Harley had also initiated specialprograms to help its dealers attract and retain cus-tomers, with notable success. Perhaps the most signifi-cant program was—and continues to be—the HarleyOwners Group (HOG), created in 1983. Begun as a way tocommunicate more effectively with the company’s endusers, HOG quickly grew into the world’s largest motor-cycle club. And dealers regained confidence that Harleycould and would be a dependable partner. At the same

136 Teerlink

HBR029ch7 8/15/01 1:37 PM Page 136

time, important improvements were made on the oper-ating side of the business, and a financial restructuringpositioned the company to go public.

By 1986, the company’s prognosis looked good. Ourmanufacturing costs were falling; reductions in in-process inventories and associated carrying costs gener-ated savings of more than $40 million a year. Our qualityhad improved; the number of warranty claims was goingdown. Our dealer network was revitalized and growing.And then we went public in July, and the offering raised$25 million more than the underwriters had expected.

Engines of Change

Once it was clear the present was in good order, it wastime to think about the future. We still had a great dealof rebuilding to do. Yes, we had a great brand, newproducts were coming to market, quality was improving,and we were making a profit. But our quality standardswere not on par with our competitors’, and our coststructure was still the highest in the industry. Internally,the company was still a shell of its former self. Wholedepartments had been hollowed out by the layoffs. Andmany other good people had left voluntarily, hoping tofind more promising futures elsewhere. Who couldblame them?

I myself didn’t have a plan for the company in myback pocket. I only knew that capturing the ideas of ourpeople—all the people at Harley—was critical to ourfuture success.

Tom Gelb, then vice president of manufacturing, JohnCampbell, vice president of human resources, and I weremeeting a lot in those days—mostly informally—to talkabout what the company was doing and why. I knew we


HBR029ch7 8/15/01 1:37 PM Page 137

needed big changes in the motorcycle division. We had toidentify some sort of strategy that could carry everyoneforward—everyone meaning employees, customers, andall other stakeholders. We had to improve operations.And I felt strongly that we needed to change the wayemployees were being treated. They could no longer beprivates, taking orders and operating within strict limits.We needed to continue to push, and push hard, to create amuch more inclusive and collegial work atmosphere.

One idea we came up with in those meetings was gainsharing—a program that would allow all employees toshare in the company’s financial success. We saw it as apossible toehold—a way to focus everyone on how wecould all get better together. But an organizational-change consultant quickly set us straight. He told us thatgain sharing would be like putting a Band-Aid on a gun-shot wound. He recommended that we talk to a few otherconsultants, get their input, and then make the call.

Enter Lee Ozley—the second consultant we spokewith, the one we hired, the one who has been with thecompany ever since, and who has become, over the years,my mentor, coach, sounding board, and good friend.Lee’s first meeting with me was typical of the kind ofbrutal candor he brought to Harley. The gain-sharingprogram, he pointed out, had been created at the top andwas about to be imposed from the top. So much forinclusive leadership!

Good Intentions

Well, we didn’t pursue the gain-sharing program.Instead, over the course of the next several months, weexplored several questions: How does constant crisismanagement affect a workforce? How can we—in theabsence of a crisis—create an environment where

138 Teerlink

HBR029ch7 8/15/01 1:37 PM Page 138

employees want to do better, where they care about thecompany on a personal level and work together toimprove both individual and overall performance?

Lee talked a lot about the psychologist AbrahamMaslow in those early months. One of Maslow’s theories,as many people know, was that absent a crisis, peoplerarely commit to a program that is imposed on them. Butthey will willingly commit to a program they help create.That thought made a huge impression on me. And even-tually, in late 1988, we agreed to launch a series of pro-grams designed to elicit ideas, thoughts, concerns, com-plaints, and vision from our employees—across alldepartments and functions.

First we approached the union leadership of theroughly 700 employees in the Wisconsin operations andasked them to help us create a vision for the business.Lee had warned us that it would take 12 to 18 months toprepare people—both union and salaried leadership—forsuch a process. But I decided we couldn’t wait. Our unioncontracts expired in three months, and I wanted a newlabor-management relationship sanctioned before then.The solution I came up with was to put in place one-yearcontracts with each local union that included the jointvision-building process. That happened.

Over the next several months, many people in theWisconsin operations worked hard to craft a joint vision.First, about 70 leaders, both union and management,wrote down their individual ideas for what Harley shouldbecome. Then a facilitator met with them in groups tobuild consensus. Finally, all 70 people met for three daysto forge a shared strategic vision for Harley-Davidson.The sense of exhilaration at the end of the process waswonderful.

Our next step was to roll the vision out to the rest ofthe company. But an interesting thing happened at the


HBR029ch7 8/15/01 1:37 PM Page 139

first presentation, where union and management leaderswere addressing a primarily salaried audience. One of thesalaried employees asked, “Who represented us in thisprocess?” I was genuinely ashamed. Once again, we hadbehaved like traditional managers. We had not includedeveryone who should have been a part of the process. Butwe quickly made amends, establishing two groups ofsalaried employees to share in the responsibility ofimplementing the vision.

Despite the great progress we made in crafting ashared vision, Lee’s warning came home to roost. Ourleaders weren’t prepared to implement the kind of dra-matic change the shared vision called for. In fact, imple-menting the vision came off as a sort of forced march,sending mixed messages and causing confusion on boththe shop floor and in the offices. Not surprisingly, thelargest local union opted out of the process after twoyears. I learned that changing people’s long-heldassumptions and behaviors takes time. That doesn’tmean you give up; it just means you keep your focus andaccept that the journey has its downs as well as its ups.Ultimately, although the joint visioning process did notsucceed as we had hoped, it did lay the groundwork forshared leadership agreements with the unions in 1997.

What Does Rich Want?

In March 1989, I became CEO, and I soon decided thatit was time to bring together the company’s 60 seniorexecutives at an off-site meeting. I wanted them to getto know one another and begin to work as a truly uni-fied team.

I discussed the idea with Lee and the executive com-mittee, and we developed a program that would feature

140 Teerlink

HBR029ch7 8/15/01 1:37 PM Page 140

two expert speakers on the issue of change and one out-side CEO who had led a major change effort. After hear-ing the presentations, we would break into groups andtalk about what we’d heard, what we thought we couldaccomplish, and how we could do it.

At the opening ceremony, I addressed the group.“We’re here to learn,” I said. “There are no right answers.

We’re here to have fun,to get to know oneanother.” And indeed,that’s what seemed tohappen—at first. Thespeakers were wellreceived, and the first

night, people did skits that were hilarious—the iceseemed to have been broken. I was feeling good aboutthe whole event.

As planned, after each speaker, we moved into smallercross-divisional, divisional, functional, and cross-func-tional groups. People in these groups would talk, and asthey did, I would visit each cluster with the speaker sothat people could ask questions. I noticed, as I went intoeach room, the question “What does Rich want?” onmany of the easels the groups were using to record theirthoughts.

I felt terrible. I thought we had been so clear at theoutset that the point was not to figure out what I wantedbut to figure out what everyone wanted for the company.The senior executives weren’t supposed to be trying toguess the right answer according to Rich. And yet,clearly, they had a concern about a hidden agenda.

At the closing session, I expressed my frustration. Iblasted the whole group. I said, “We need a revolution,and I’m not going to lead it. You need to lead it.” I told

I noticed, as I went into each room, the question “What does Rich want?” on many ofthe easels the groups were using to record their thoughts.


HBR029ch7 8/15/01 1:37 PM Page 141

them I was disappointed with what I’d seen on the easels.That ended the meeting.

When I returned to the office on Monday, one of mycolleagues was waiting for me. She took me aside and letme have it. She said, “You know, you really stuck a pin inour balloon at that session. We were doing fine. We weregoing through a natural process, getting to know oneanother, expressing our fears as well as our thoughtsabout what we wanted for the company.”

And then I realized that what had been upsetting memost was simply that the participants had not done whatI wanted them to do. My ego got in the way. Rather thanbeing happy that people were talking, and understandingthat it was normal for people to want to figure outwhether “the boss” had an agenda, I had, in my heatedenthusiasm for inclusivity, picked up on—and pickedat—what was in reality a minor point. My expectationsweren’t important at that exercise. What was importantwere their expectations. And they were happy. Too fre-quently, we as leaders are trying to satisfy ourselvesrather than others.

Fortunately, that colleague felt comfortable enough totell me that I had been wrong. Lee and I talked throughthe event and discussed why we both felt responsible.

We realized, however,we had introducedsomething that wouldbe useful in the future.And I realized that Ihad to get myself recali-brated if these meetings

were going to be successful. I did. The second sessionwent well. By the third, we had jettisoned our facilita-tors—as had been the plan—and took responsibility toteach one another instead of listening to the experts.

When you’re a CEO, there is always a barrier between you and the rest of thecompany, no matter how hardyou try to break it down.

142 Teerlink

HBR029ch7 8/15/01 1:37 PM Page 142

Ongoing Process

I retired as Harley’s CEO in 1999, and I now serve on itsboard. Looking back, I can see that many times duringthe course of my tenure, I slipped in my commitment toinclusive leadership—even as I was preaching its virtues.Each time (or most times—I imagine that there weretimes when no one called me on it; I hope there weren’ttoo many) Lee or another colleague pulled me up shortand reminded me that I wasn’t “walking the talk.” I wasalways grateful.

When you’re a CEO, there is always a barrier betweenyou and the rest of the company, no matter how hardyou try to break it down. And so I also tried very hard tolearn to discipline myself—to step out of myself before Ispoke or took action and ensure that I wasn’t going toslip out of my chosen role.

Overall, I believe I succeeded in my journey to be a dif-ferent kind of leader. Harley’s culture has changed, butthe work is not done. Transforming a culture takes time.Everybody hasn’t fully bought in to the inclusiveapproach. We still have some people who think theyknow all the answers, but these people are getting fewerin number. We still have people who just want to bringtheir bodies and not their whole selves, mind included, towork. But their ranks are dwindling, too. We’ve beentransforming ourselves since the buyout and will still beat it ten years from now. It is a journey that will neverend unless we let it.

Originally published in July–August 2000Reprint R00411


HBR029ch7 8/15/01 1:37 PM Page 143

Waking Up IBMHow a Gang of Unlikely RebelsTransformed Big Blue

Executive Summary

IN THE EARLY 1990S, IBM was a has-been. Fujitsu,Digital Equipment, and Compaq were hammering downits hardware margins. EDS and Andersen Consultingwere stealing the hearts of CIOs. Intel and Microsoftwere running away with PC profits.

Today, Big Blue is back on top, a leader in e-business services. This is the story of how the company,which had lagged behind every computer trend sincethe mainframe, caught the Internet wave. Much of thecredit for the turnaround goes to a small band ofactivists who built a bonfire under IBM’s rather broadbehind.

It stared in February 1994, when a lone midlevelIBM programmer watched Sun Microsystems pirateIBM’s Winter Olympics data for its own rogue Web site.Dave Grossman knew that IBM’s muckety-mucks were

145

HBR029ch8 8/15/01 1:38 PM Page 145

clueless about the Web. But he was convinced that ifnothing changed Sun would eat Big Blue’s lunch.

Frustrated in his attempts to warn executives over thephone, he drove down to Armonk, walked straight intoheadquarters with a Unix workstation in his arms, set it upin a closet, and demonstrated the future of computing toa trio of IBM execs. One of them was John Patrick, headof marketing for the hugely successful ThinkPad, whoquickly became his mentor. Together, building simultane-ously from the top and the bottom of the organizationthrough an ever-widening grassroots coalition of techni-cians and executives, they put IBM on the Web andmorphed it into an e-business powerhouse. People whowant to foment similarly successful insurrections can learna lot from their example.

D IBM was a case study incomplacency? Insulated from the real world by layerupon layer of dutiful managers and obsequious staff,IBM’s executives were too busy fighting their endless turf

battles to notice that thecompany’s once unas-sailable leadership posi-tion was crumblingaround them. The com-pany that held the topspot on Fortune’s list ofmost admired corpora-tions for four years run-

ning in the mid-1980s was in dire need of saving by theearly 1990s. Fujitsu, Digital Equipment, and Compaqwere hammering down hardware margins. EDS and

How did a company that had lagged behind every computer trend since the mainframe catch the Internet wave—a wave that even Bill Gates and Microsoft originally missed?

146 Hamel

HBR029ch8 8/15/01 1:38 PM Page 146

Andersen Consulting were stealing the hearts of CIOs.Intel and Microsoft were running away with PC profits.Customers were bemoaning the company’s arrogance. Bythe end of 1994, Lou Gerstner’s first full year as CEO, thecompany had racked up $15 billion in cumulative lossesover the previous three years, and its market cap hadplummeted from a high of $105 billion to $32 billion.Armchair consultants were nearly unanimous in theirview: Big Blue should be broken up.

Despite Gerstner’s early assertion that IBM didn’tneed a strategy (the last thing he wanted was to startanother corporatewide talk fest), IBM was rudderless ingale force winds. Yet over the next six years, the com-pany transformed itself from a besieged box maker to adominant service provider. Its Global Services unit, oncea backwater, grew into a $30 billion business with morethan 135,000 employees, and corporations flocked toIBM consultants for help in capitalizing on the Internet.By the end of 1998, IBM had completed 18,000 e-businessconsulting engagements, and about a quarter of its $82billion in revenues was Net related.

How did a company that had lagged behind everycomputer trend since the mainframe catch the Internetwave—a wave that even Bill Gates and Microsoft origi-nally missed? Much of the credit goes to a small band ofactivists who built a bonfire under IBM’s rather broadbehind. This is their story.

Missing an Olympic Opportunity

The first match was struck in 1994 in the backwoods ofIBM’s empire, on a hilltop in Ithaca, New York, by a typi-cal self-absorbed programmer. David Grossman was amidlevel IBMer stationed at Cornell University’s Theory

Waking Up IBM 147

HBR029ch8 8/15/01 1:38 PM Page 147

Center, a nondescript building hidden away in the south-east corner of the engineering quad. Using a supercom-puter connected to an early version of the Internet,Grossman was one of the first people in the world todownload the Mosaic browser and experience the graph-ical world of the Web. Grossman’s fecund imaginationquickly conjured up a wealth of interesting applicationsfor the nascent technology. But it was an event in Febru-ary, as snow dusted the ground around the Theory Cen-ter, that hardened his determination to help get IBM outin front of what he knew would at the very least be theNext Big Thing—and might very well be the Ultimate Big Thing.

The Winter Olympics had just started in Lillehammer,Norway, and IBM was its official technology sponsor,responsible for collecting and displaying all the results.Watching the games at home, Grossman saw the IBMlogo on the bottom of his TV screen and sat through thefeel-good ads touting IBM’s contribution to the event.But when he sat in front of his UNIX workstation andsurfed the Web, he got a totally different picture. A rogueOlympics Web site, run by Sun Microsystems, was takingIBM’s raw data feed and presenting it under the Sun ban-ner. “If I didn’t know any better,” says Grossman, “Iwould have thought that the data was being provided bySun. And IBM didn’t have a clue as to what was happen-ing on the open Internet. It bothered me.”

The fact that IBM’s muckety-mucks were cluelessabout the Web wasn’t exactly news to Grossman. Whenhe had landed at IBM a few years earlier, everyone wasstill using mainframe terminals. “I was shocked,” heremembers. “I came from a progressive computing envi-ronment and was telling people at IBM that there wasthis thing called UNIX—there was an Internet. No oneknew what I was talking about.”

148 Hamel

HBR029ch8 8/15/01 1:38 PM Page 148

This time, though, he felt embarrassed for IBM, andhe was irked. He logged on to the corporate directoryand looked up the name of the senior executive in chargeof all IBM marketing, Abby Kohnstamm. Then he senther a message informing her that IBM’s Olympic feed

was being ripped off. Afew days later, one of herminions working in Lille-hammer called Grossmanback. At the end of a frus-trating conversation,Grossman had the feeling

that one of them was living on another planet. Ever per-sistent, Grossman tried to send some screen shots fromSun’s Web site to IBM’s marketing staff in Lillehammer,but IBM’s internal e-mail system couldn’t cope with theWeb software. That didn’t stop IBM’s diligent legaldepartment from sending Sun a cease-and-desist letter,which succeeded in shutting down the site.

Most frontline employees would have left it at that.But Grossman felt IBM was missing a bigger point: Sunwas about to eat Big Blue’s lunch. After everyone hadcome back from the Olympics, he drove down to IBM’sheadquarters, four hours away in Armonk, New York, topersonally show Kohnstamm the Internet.

A Virtual Team Takes Shape

When he arrived, Grossman walked in unattended, aUNIX workstation in his arms. Wearing a programmer’suniform of khakis and an open-necked shirt, he woundhis way up to the third floor—the sanctum sanctorum ofthe largest computer company in the world. Borrowing aT1 line from someone who had been working on a videoproject, he strung it down the hall to a storage closet

As sober-suited IBM executives scurried through their rounds, Mick Jagger could be heard wailing from the closet.

Waking Up IBM 149

HBR029ch8 8/15/01 1:38 PM Page 149

where he plugged it into the back of his workstation. Hewas now ready for his demo—a tour of some early Websites, including one for the Rolling Stones. As sober-suited IBM executives scurried through their rounds,Mick Jagger could be heard wailing from the closet.

In addition to Kohnstamm, two others were presentat that first demo. One was Irving Wladawsky-Berger,head of the supercomputer division where Grossmanworked. The other was John Patrick, who sat on a strat-egy task force with Wladawsky-Berger. Patrick, a careerIBMer and lifelong gadget freak, had been head of mar-keting for the hugely successful ThinkPad laptop com-puter and was working in corporate strategy, scoutingfor his next big project. Within minutes, Grossman hadhis full attention. “When I saw the Web for the firsttime,” says Patrick, “all the bells and whistles went off. Itsability to include colorful, interesting graphics and tolink to audio and video content blew my mind.”

Not everyone saw what Patrick saw in that primitivefirst browser. “Two people can see the same thing buthave a very different understanding of the implications,”he recalls. “A lot of people [would say], ‘What’s the bigdeal about the Web?’ But I could see that people woulddo their banking here and get access to all kinds of infor-mation. I had been using on-line systems like Com-puServe for a long time. For people who weren’t alreadyusing on-line systems, it was harder for them to see.”

Their passions fueled by the Web’s limitless possibili-ties, Patrick and Grossman became IBM’s Internet tagteam, with Patrick doing the business translation forGrossman and Grossman doing the technology transla-tion for Patrick. Patrick acted as a sponsor and aresource broker. Grossman developed intimate linkswith the Net-heads in IBM’s far-flung development com-

150 Hamel

HBR029ch8 8/15/01 1:38 PM Page 150

munity. “The hardest part for people on the street likeme,” says Grossman, “was how to get senior-level atten-tion within IBM.” Patrick became his mentor and his go-between.

After seeing Grossman’s demo, Patrick hired him, andthey soon hooked up with another Internet activistwithin IBM, David Singer. Singer was a researcher in

Alameda, California,who had written oneof the first Gopherprograms, whichfetched informationoff the Net. Grossmanand Singer startedbuilding a primitivecorporate intranet,

and Patrick published a nine-page manifesto extollingthe Web. Entitled “Get Connected,” the manifesto out-lined six ways IBM could leverage the Web:

1. Replace paper communications with e-mail.

2. Give every employee an e-mail address.

3. Make top executives available to customers andinvestors on-line.

4. Build a home page to better communicate with cus-tomers.

5. Print a Web address on everything, and put all mar-keting on-line.

6. Use the home page for e-commerce.

The Get Connected paper, distributed informally by e-mail, found a ready audience among IBM’s unheralded

Like dissidents using a purloined duplicator in the old Soviet Union, Patrick and Grossman used the Web to build a community of Web fans that would ultimatelytransform IBM.

Waking Up IBM 151

HBR029ch8 8/15/01 1:38 PM Page 151

Internet aficionados. The next step was to set up an on-line news group of the sort that allowed IBM’s under-ground hackers to trade technical tidbits. “Very fewpeople higher up even knew this stuff existed,” says Gross-man. Within months, more than 300 enthusiasts wouldjoin the virtual Get Connected team. Like dissidents usinga purloined duplicator in the old Soviet Union, Patrickand Grossman would use the Web to build a communityof Web fans that would ultimately transform IBM.

As Patrick’s group began to blossom, some arguedthat he should “go corporate” and turn the nascent Webinitiative into an officially sanctioned project. Patrick’sboss, senior VP for strategy and development JimCanavino, disagreed. “You know, we could set up somesort of department and give you a title,” Canavinoremarked to Patrick, “but I think that would be a badidea. Try to keep this grassroots thing going as long aspossible.” Patrick needed to infiltrate IBM rather thanmanage some splendidly isolated project team. It wouldbe easy for others at IBM to ignore a dinky department,but they couldn’t stand in the way of a groundswell.

Still, Canavino wasn’t above using his role as head ofstrategy to give the fledgling initiative a push. To avoidthe danger of going quickly from having no IBM Web siteto having dozens of uncoordinated ones, Canavinodecreed that nobody could build a site without Patrick’sapproval. Though few in IBM had any inkling of what theInternet would become, Patrick had become IBM’s semi-official Internet czar.

“Where’s the Buy Button?”

Patrick’s volunteer army was a widely dispersed group ofNet addicts, many of whom had no idea that othersshared their passion. “What John ended up providing,”

152 Hamel

HBR029ch8 8/15/01 1:38 PM Page 152

says Grossman, “was the ability to articulate and sum-marize what everyone was doing and to open a lot ofdoors.” In turn, the Net-heads introduced Patrick to theculture of the Internet, with its egalitarian ideals andtrial-by-fire approach to developing new technologies.When the Get Connected conspirators gathered for theirfirst physical meeting, remembers Grossman, “the ques-tion on everybody’s lips was, How do we wake this com-pany up?” (For more on this, see “How to Start an Insur-rection” at the end of this article.)

Patrick gathered a small group of his Get Connectedrenegades, including Grossman, at his vacation house,set deep in the woods of western Pennsylvania. Therethey cobbled together a mock-up of an IBM home page.The next step was to get through to Gerstner’s personaltechnology adviser, who agreed to make him availablefor a demo of the prospective IBM corporate Web site.When Gerstner saw the mock-up, his first question was,“Where’s the buy button?” Gerstner wasn’t a quickstudy—he was an instant study. But Grossman andPatrick knew that an intrigued CEO wasn’t enough.There were thousands of others who still needed to getthe Internet religion.

Their first chance for a mass conversion came at ameeting of IBM’s top 300 officers on May 11, 1994. Havingschemed to get himself on the agenda, Patrick drovehis point home hard. He started by showing IBM’s top brass some other sites that were already up and running,including ones done by Hewlett-Packard; Sun Microsys-tems; the Red Sage restaurant in Washington, D.C.; andGrossman’s six-year-old son Andrew. The point was clear:on the Web, everyone could have a virtual presence.

Patrick ended the demo by saying, “Oh, by the way,IBM is going to have a home page too, and this is what itwill look like.” He showed the startled executives a

Waking Up IBM 153

HBR029ch8 8/15/01 1:38 PM Page 153

mock-up of www.ibm.com, complete with a 36.2-secondvideo clip of Gerstner saying, “My name is Lou Gerstner.Welcome to IBM.”

Still, many IBM old-timers remained skeptical. RecallsPatrick: “A lot of people were saying, ‘How do you makemoney at this?’ I said, ‘I have no idea. All I know is thatthis is the most powerful, important form of communi-cation both inside and outside the company that hasever existed.’”

Shortly after the May meeting, Patrick and a few col-leagues showed up at one of the first Internet Worldtrade conventions. The star of the show, with the biggestbooth, was rival Digital Equipment. Like Grossman’sbefore him, Patrick’s competitive fires were stoked. Thenext day, the convention’s organizers auctioned off spacefor the next show, scheduled for December, and Patricksigned IBM up for the biggest display, at a cost of tens ofthousands of dollars. “It was money I did not have,”admits Patrick, “but I knew I could find it somehow. Ifyou don’t occasionally exceed your formal authority, youare not pushing the envelope.”

Now that IBM’s name was on the line, Patrick had arallying point for all of the company’s various Internet-related projects. Here was his chance to seed his messageacross the entire company. He sent letters to the generalmanagers of all the business units asking for anythingthey had that smelled like the Internet. They would haveto put in only a little money, he promised, and he wouldcoordinate everything. It turned out that IBM had a lotmore Web technology brewing than even he hadexpected. But none of it was really ready to go to market.Still, by December, Patrick was able to showcase IBM’sGlobal Network, as the world’s largest Internet serviceprovider, as well as a Web browser that preceded both

154 Hamel

HBR029ch8 8/15/01 1:38 PM Page 154

Netscape’s Navigator and Microsoft’s Internet Explorer.IBM stole the show and became a fixture at every Inter-net World thereafter.

Constantly fighting IBM’s parochialism, Patrick tookevery opportunity to drive home the point that the Webwas a companywide issue and not the preserve of a singledivision. At the next Internet World, in June 1995, hechallenged his compatriots to leave their local biases atthe door: “The night before the show, I got everybodytogether in an auditorium and said, ‘We are here becausewe are the IBM Internet team for the next three days.You are not IBM Austin or IBM Germany.’ That is part ofthe culture of the Internet—boundaryless, flat.”

The huge IBM booth generated a lot of curiosityamong the show’s other participants. When people askedPatrick to whom he reported, he said, “The Internet.”When they asked him about his organization, he replied,“You’re looking at it, and there are hundreds more.”

Throwing Hand Grenades

Patrick was a relentless campaigner, spreading the goodword about the Internet in countless speeches inside andoutside IBM. “Somebody would invite me to talk aboutthe ThinkPad,” he recalls, “and I would come talk aboutthe Internet instead. I’d use the ThinkPad to bring upWeb page presentations rather than PowerPoint slides.”He also made himself accessible to the media. But evenwhen talking to reporters, his prime constituency wasstill the vast swath of unconverted IBMers. He justcouldn’t shut up about the Internet. Says Patrick: “If youbelieve it, you’ve got to be out there constantly talkingabout it, not sometimes, but all the time. If you knowyou’re right, you just keep going.”

Waking Up IBM 155

HBR029ch8 8/15/01 1:38 PM Page 155

While Patrick and his crew were throwing Internethand grenades into every meeting they could wheedletheir way into, Gerstner was fanning the flames fromabove. Gerstner’s early belief in the importance of net-work computing dovetailed nicely with the logic of theInternet. Having bought into Patrick’s pitch, Gerstnerwas ever ready to give IBM’s Web-heads a boost. Heinsisted that the company put its annual and quarterlyreports on the Web, and he signed up to give a keynoteaddress at Internet World. This was while Bill Gates andothers were still dissing the Web as an insecure mediumfor consumer e-commerce. Within IBM, Patrick becamea trusted emissary between the company’s buttoned-down corporate types and the T-shirted buccaneers whowere plugged into Net culture and living on Internettime. Patrick had the ear of IBM’s aristocracy, and hismessage was simple and unequivocal: “Miss this and youmiss the future of computing.” At the same time, Patrickconvinced Grossman and his ilk that not everyone in thehead office was a Neanderthal. “I used to think that IBMat senior levels was clueless, that these guys had no ideahow to run a company,” says Grossman. “But one of themany things that has impressed me is that the peoplewho are running this company are really brilliant busi-nesspeople. Somehow we connected them to the street.Knowing how to shorten paths to those decision makerswas key.”

When IBM finally set up a small, formal Internetgroup, with Patrick as chief technical officer, he insistedthat the team stay separate from IBM’s traditional soft-ware development organization. His logic: “I do believethere’s a benefit in being separate. Otherwise, we’d haveto start going to meetings. Pretty soon we’d be part ofsomeone else’s organization, and a budget cut wouldcome along, and we’d be gone.”

156 Hamel

HBR029ch8 8/15/01 1:38 PM Page 156

Even with the formal unit in place, Patrick and Gross-man didn’t disband their grassroots coalition. As the1996 Summer Olympics approached, the group wentthrough several watershed events. Patrick lent Grossmanout for 18 months to corporate marketing, which was incharge of the Olympics project. For the first time, theOlympics would have an official Web site, and IBMwould build it. Grossman launched himself into buildingthe site and was soon begging Patrick for extra bodies.“Patrick did the magic to get them hired,” says Gross-man, “and I morphed from doing the grunt technicalwork to being Tom Sawyer and getting other people tohelp whitewash the fence.”

To prepare for the Olympics, Grossman and his teamhad also started developing Web sites for other sportingevents such as the 1995 U.S. Open and Wimbledon. Forthe U.S. Open site, he gave a couple of college internsfrom MIT the task of writing a program to connect ascoring database to the Web site. “By the end of the sum-mer,” remembers Grossman, “we were sitting in a trailer,barely keeping together a Web site with a million peoplea day pounding away at it for scores. It was held togetherby Scotch tape, but we were learning about scalability.” Itwas amazing, thought Grossman, that all those peoplewould come to a site merely for sports scores.

IBM’s second surprise came in 1996 when a chessmatch between world champion Garry Kasparov and anIBM supercomputer named Deep Blue generated a floodof global interest. Corporate marketing had asked Gross-man earlier to build the Web site for the match, but hewas booked with too many other assignments, so the sitewas outsourced to an advertising agency that did littlemore than put up a cheesy chessboard. The day of thefirst match, the site was overloaded with traffic andcrashed.

Waking Up IBM 157

HBR029ch8 8/15/01 1:38 PM Page 157

“Nobody had any idea this was going to be such a bigdeal,” says Patrick. IBM went into panic mode. Grossmanand a handful of IBM’s best Web engineers jumped in totake over the site. With only 36 hours to revamp it beforethe next match, they got Wladawsky-Berger to pull a$500,000 supercomputer off the assembly line. The sitedidn’t crash again, but the incident raised the anxietylevel about the upcoming Olympics. If IBM was havingdifficulty running a Web site for a chess match, whatwere the Olympics going to be like? The incident suc-ceeded in convincing a few more skeptics that the Inter-net was going to be beyond Big.

The Olympics site had to be able to withstand any-thing. Patrick went tin-cupping again, asking all the gen-eral managers to lend him their best people and bestequipment. He got not one supercomputer, but three,and his team grew to about 100 people. By the time itwas over, IBM had built what was then the world’slargest Web site, which withstood up to 17 million hits aday with few shutdowns. The content on the site wasreplicated on servers across four continents. IBM evenlearned how to do a little e-commerce when a demo sitefor on-line ticket sales attracted a flood of credit cardnumbers and $5 million in orders.

The Power of Results

For Patrick and Grossman, the Olympics was just onemore high-profile way to show IBM the possibilities ofthe Internet. It was also an easy way to get funding fordevelopment. “I used the Olympics as a front,” admitsGrossman. “What I was doing, without telling anyone,was getting computing resources. I also thought thefastest way to get IBM to change was to work from the

158 Hamel

HBR029ch8 8/15/01 1:38 PM Page 158

outside in. If IBM saw itself written about in the papers,then it would change faster than if we got mired in aninternal process.”

Grossman’s on-the-fly development, in public no less,was the complete antithesis of IBM’s traditional way ofdoing things, which was to push developers to perfectproducts before letting them out the door. It was the dif-ference between improv comedy and a carefullyrehearsed Broadway play. The old model didn’t makemuch sense on the Web, where if something breaks, youcan fix it without sending out millions of CD-ROMs withnew software. You just change the software on theserver, and everyone who logs on automatically gets thenew version.

Grossman and Patrick quickly concluded that creat-ing Web-enabled software called for a new set of soft-ware development principles, which they summarizedand shared within the burgeoning IBM Web community:

• Start simple; grow fast.

• Trial by fire.

• Just don’t inhale (the stale air of orthodoxy).

• Just enough is good enough.

• Skip the krill (go to the top of the food chain whenyou’re trying to sell your idea).

• Wherever you go, there you are (the Net has nobounds).

• No blinders.

• Take risks; make mistakes quickly; fix them fast.

• Don’t get pinned down (to any one way of thinking).

Waking Up IBM 159

HBR029ch8 8/15/01 1:38 PM Page 159

Much of the technology that Grossman and his crewfirst prototyped would later make its way into industrial-strength products. For instance, the Web server softwaredeveloped for the Olympics evolved into a product calledWebsphere, and much of what Grossman’s grouplearned formed the basis for a Web-hosting business thattoday supports tens of thousands of Web sites.

Following the Olympics, the Internet group steppedup its proselytizing within IBM. Grossman, who hadbecome the senior technical staff member on Patrick’steam, set up an Internet lab to bring in executives from

all over the company toexperience the Web’spossibilities. Patrick’sgroup also started aproject called “WebAhead,” which workedto revolutionize thecompany’s own IT sys-

tems through Internet technology. For instance, theteam took the old terminal-based corporate directoryand wrote a Java application that gave it a great graphi-cal interface and cool features. With a few clicks,employees could look up a colleague, see what computerskills he or she had, and then ask the directory to listevery other employee at IBM with those same skills.These “Blue Pages” were an instant hit.

Only a few dozen people officially worked for theInternet group, so Patrick was constantly pleading toborrow people (who were usually already part of his vir-tual team) from other departments. In this effort, hismost important ally was the team’s ever-lengthening listof success stories. People could argue with positionpapers, but they couldn’t argue with results. “I have

“We have never been a threat to any other part of the company. From the beginning, our goal was to help IBM become the Internet Business Machines company.”

160 Hamel

HBR029ch8 8/15/01 1:38 PM Page 160

never been turned down on anything I have asked for,and I have asked for a lot,” he says. “I would go to a gen-eral manager and say, ‘I need you to pull some disk drivesfrom the assembly line, and I need your top engineer.What you will get out of it is unique. Your guy is going tocome back to your group, and you are going to have ahell of a reference story to talk about. It will be great PR.We will make your stuff work on the Internet.’” Patrickhad gained credibility without a big job title or a mega-budget.

Patrick was hard to refuse, partly because it was clearthat he was operating in IBM’s interests as a whole andnot just fighting for his own little group. As he puts it, “Ididn’t have any allegiance to any one product group.Although I had a budget that came out of the softwaregroup, I didn’t think of us as part of the software group.When somebody called us and asked for help, we didn’task them for a budget code. We’d say, ‘Sure.’ We havenever been a threat to any other part of the company.From the beginning, our goal was to help IBM becomethe Internet Business Machines company.”

Patrick was quick to assure would-be donors that therelationships he was forging worked both ways. Hewould borrow people from various business units, but atany given time, about a quarter of his own people wouldbe out on loan to other units, and Web Ahead alumniwere regularly posted to permanent positions acrossIBM. When that happened, he would tell his remainingstaff, “We did not lose Bill. We colonized the networkhardware division. Now there is one of us living there.”

Again and again, throughout their Internet campaign,Patrick and Grossman broke long-standing IBM rulesand overstepped the boundaries of their own authority.But because their cause was so important and their

Waking Up IBM 161

HBR029ch8 8/15/01 1:38 PM Page 161

commitment to IBM’s success so visibly selfless, they gotaway with things that had often sunk careers at Big Blue.Then and now, Patrick is unapologetic: “If you think ofyourself as being in a box with boundaries, you’re notgoing to have any breakthroughs. If [people on my team]come to me and say, ‘We failed because we didn’t havethe authority to do something,’ I’ll say that’s crazy.”

Inside IBM and out, Patrick and Grossman are todayrecognized for their pivotal contribution to their com-pany’s e-business metamorphosis. With the support of aprochange CEO, these two unlikely heroes—a softwarenerd and a corporate staffer—helped IBM do somethingit hadn’t done for a couple of decades: lead from thefront.

How to Start an Insurrection

Is it clear to you that your company needs to be shakenup? Then it’s time you became a revolutionary. Here areseven steps for organizing a corporate insurrection.

1. Establish a point of view. In a world of people whostand for nothing more than more of the same, a sharplyarticulated POV is your greatest asset. It’s a sword that letsyou slay the dragons of precedent. It’s a rudder that letsyou steer a steady course when others are blown aboutby fad and whim. And it’s a beacon that attracts thosewho are looking for something worthy of their allegiance.A powerful POV is credible, coherent, compelling, andcommercial. To be credible, it must be founded on unim-peachable data. To be coherent, it must be logical, layingout a bulletproof argument. To be compelling, it must

162 Hamel

HBR029ch8 8/15/01 1:38 PM Page 162

speak to people’s emotions, telling them why your causewill make a difference in the world. To be commercial, itmust have a clear link to the bottom line.

2. Write a manifesto. It’s not enough to have an ideology;you have to be able to pass it on, to infect others withyour ideas. Like Thomas Paine, whose Common Sensebecame the inspiration for the American Revolution, youhave to write a manifesto. It doesn’t have to be long, butit must capture people’s imaginations. It must paint a pic-ture of what is and what is coming that causes discom-fort. And it must provide a vision of what could be thatinspires hope.

3. Create a coalition. You can’t change the direction ofyour company all by yourself. You need to build a coali-tion, a group of colleagues who share your vision andpassion. It’s easy to dismiss corporate rebels when theyare fragmented and isolated. But when they presentthemselves as a coordinated group, speaking in a singlevoice, they cannot be ignored. And remember, as youstruggle to attract recruits to your cause, you will have anadvantage over top management. Your army will bemade up of volunteers; theirs will be composed ofconscripts. Conscripts fight to stay alive; volunteers fightto win.

4. Pick your targets. Sooner or later, a manifesto has tobecome a mandate if it’s going to make a difference.The movement has to get the blessing of the suits.That’s why activists always identify and target a poten-tial champion—someone or a group of someones thatcan yank the real levers of power. Ultimately, the sup-port of senior management is the object of your cru-sade. Make an effort to understand them—the pressuresthey face, the objectives they have to fulfill. Find some

Waking Up IBM 163

HBR029ch8 8/15/01 1:38 PM Page 163

who are searching for help and ideas, and go afterthem. If necessary, bend your ideals a bit to fit theirgoals. And don’t forget that leaders are often morereceptive to new thinking than are the minions whoserve them.

5. Co-opt and neutralize. Some activists further their causesby confronting and embarrassing their adversaries. Suchtactics may work in the public sphere, but in a business set-ting they’ll probably get you fired. You need to disarmand co-opt, not demean and humiliate. To win over IBM’sfeudal lords, John Patrick constructed a set of win-winpropositions for them: Lend me some talent, and I’ll build ashowcase for your products. Let me borrow a few of yourtop people, and I’ll send them back with prototypes ofcool new products. Reciprocity wins converts; rantingleaves you isolated and powerless.

6. Find a translator. Imagine how a buttoned-down dadlooks at a daughter who comes home with green hairand an eyebrow ring. That’s the way top management islikely to view you and your coconspirators. And that’swhy you need a translator, someone who can build abridge between you and the people with the power. AtIBM, Patrick was a translator for Dave Grossman. Hehelped the top brass understand the connection betweenthe apparent chaos of the Web and the disciplinedworld of large-scale corporate computing. Senior staffersand newly appointed executives are often good transla-tor candidates—they’re usually hungry for an agenda tocall their own.

7. Win small, win early, win often. None of your organiz-ing efforts is worth anything if you can’t demonstrate thatyour ideas actually work. You need results. Start small.Unless you harbor kamikaze instincts, search for demon-

164 Hamel

HBR029ch8 8/15/01 1:38 PM Page 164

stration projects that won’t sink you or your cause if theyshould fail—for some of them will fail. You may have toput together a string of successful projects before topmanagement starts throwing money your way. You haveto help your company feel its way toward revolutionaryopportunities, step by step. And as your record of winsgets longer, you’ll find it much easier to make the transi-tion from an isolated initiative to an integral part of thebusiness. Not only will you have won the battles, you willhave won the war.

Originally published in July–August 2000Reprint R00406

The author acknowledges the assistance of Erick Schonfeld in the prepa-ration of this article.

Waking Up IBM 165

HBR029ch8 8/15/01 1:38 PM Page 165

167

About the Contributors

is Cahners-Rabb Professor of BusinessAdministration at Harvard Business School, where he teachesin the areas of organization effectiveness, human resourcemanagement, and organizational change. Prior to joining theHarvard faculty, he was Director of Organization Research andDevelopment at Corning, Inc., where he was responsible forstimulating a number of innovations in management. He hasauthored or coauthored several books and many articles. TheCritical Path to Corporate Renewal, which deals with the prob-lem of large-scale corporate change, won the Johnson, Smith,and Knisley Award for the best book on executive leadership in1991 and was a finalist for the Academy of Management BookAward that year. In the last several years, he has developed andresearched a process by which top teams can manage strategicchange in a way that integrates their business and and organi-zational capability building objectives as advocated in thearticle he coauthored for this volume. He is Cofounder andChairman of the Center for Organizational Fitness, whichworks with top teams who want to learn how to manage sucha change process. Professor Beer has served on the editorialboard of several journals and the board of governors of theAcademy of Management, and has consulted with many For-tune 500 companies. His most recent book, Breaking the Codeof Change, deals with many of the issues raised in the article,“Cracking the Code of Change,” included in this volume.

HBR029ABO 8/15/01 1:39 PM Page 167

is the coauthor of Built to Last: SuccessfulHabits of Visionary Companies and the author of Good toGreat: Why Some Companies Make the Leap . . . And OthersDon’t. He operates a management research and teaching labo-ratory in Boulder, Colorado.

has consulted with CEOs and executives atsuch companies as Allstate, Sears, and Hughes Space andCommunications. She now leads dialogues in national policyforums at the Aspen Institute and for the California Environ-mental Dialogue, a group of more than twenty-five energycompanies, automakers, high-tech companies, and govern-ment and environmental organizations working on the state’senvironmental policy. She lives in Austin, Texas.

is Founder and Chairman of Strategos, a com-pany dedicated to helping its clients develop revolutionarystrategies. He is also Visiting Professor of Strategic and Inter-national Management at London Business School. Hamel hasled initiatives within many of the world’s leading companies.In his work he helps companies first to imagine and then tocreate the new rules, new businesses, and new industries thatwill define the industrial landscape of the future. He has pub-lished numerous articles in the Harvard Business Review, For-tune, Sloan Management Review, and The Wall Street Journaland has introduced such breakthrough concepts as strategicintent, core competence, corporate imagination, expedi-tionary marketing, and strategy as stretch. His book, Compet-ing for the Future, with C.K. Prahalad has been hailed by manyjournals as one of the decade’s most influential businessbooks, and by Business Week as “Best Management Book ofthe Year.” His most recent book, Leading the Revolution, waspublished in September 2000 and was Amazon.com’s pick asBusiness Book of the Year.

168 About the Contributors

HBR029ABO 8/15/01 1:39 PM Page 168

is the Editorial Director at HarvardBusiness School Press and a contributing editor to theHarvard Business Review.

is Professor of Pediatrics and Anesthesia atDuke University Medical Center, Chief of Critical Care, andthe former Chief Medical Director for the Duke Children’sHospital. He is board certified in Pediatrics, Pediatric Cardiol-ogy, and Pediatric Critical Care Medicine. As Chief Executiveof PracticingSmarter Inc., an organization credited with pro-viding proven enterprise-wide performance managementsolutions for the healthcare industry, Dr. Meliones is respon-sible for setting the vision and direction of the company andestablishing the clinical foundation for its performance man-agement products and services. He has served on the boardsof various academic journals and medical societies and hasheld leadership positions for these, including President andSenior Editor. In addition to numerous awards for accom-plishment in the academic arena, Dr. Meliones has receivedand continues to receive recognition for his creativity, innova-tion, and performance results in personalizing and applyingthe balanced scorecard to healthcare management. He is rec-ognized by Dr. Robert Kaplan, Balanced Scorecard Cofounderwith Dr. David Norton, as an industry leader and the onlyproven implementer of the balanced scorecard in the health-care industry.

, now an independent consultant, was asenior advisor to CSC Index. He lives in Portland, Oregon.

is the Richard P. Chapman Professor of Busi-ness Administration and Chairman of the OrganizationalBehavior Unit at the Harvard Business School. His researchinterests center on leadership and organizational change. His


HBR029ABO 8/15/01 1:39 PM Page 169

latest book, The Arc of Ambition, coauthored with JimChampy, examines the role of ambition in the making (andbreaking) of great achievers. He is currently working on abook entitled Changing Fortunes, which examines how thisvibrant economic sector came to be called the “old economy”and what its future prospects are. Professor Nohria has writ-ten several other critically acclaimed books including, Break-ing the Code of Change, Fast Forward, Beyond the Hype, Build-ing the Information Organization, and The DifferentiatedNetwork, which won the 1998 George R. Terry Award. Profes-sor Nohria lectures to corporate audiences around the globeand serves on the advisory boards of several firms. He hasbeen interviewed by ABC, CNN, and NPR, and cited frequentlyin Business Week, The Economist, Financial Times, Fortune,New York Times, and The Wall Street Journal.

At the time this article was originally published, -

was Director of Football Operations for the New YorkJets. He has served as head coach of three professional foot-ball teams: the New York Giants (1983–1990), the New Eng-land Patriots (1993–1997), and the Jets (1997–1999).

is a writer, a leading business consul-tant worldwide, and an associate fellow of Templeton Col-lege, Oxford University, England. He is the coauthor (withMark Millemann and Linda Gioja) of Surfing the Edge ofChaos. Dr. Pascale also authored the 1990 book, Managingon the Edge, and coauthored the 1981 best-selling book, TheArt of Japanese Management, with Anthony Athos. He wason the faculty of Stanford Business School for twenty years,and has written numerous articles for leading business man-agement publications.

retired as Chairman and CEO of Harley-Davidson in 1999. During his eighteen years with Harley, he

170 About the Contributors

HBR029ABO 8/15/01 1:39 PM Page 170

net30

served as Chief Financial Officer, President, Chief ExecutiveOfficer, and Chairman of the Board. He was appointed to theboard of directors in 1982 and was part of the leadership teamthat led the revitalization of the company. In 1986, he helpedto guide Harley back to public ownership. Prior to joiningHarley, he had the benefit of working at large and small, pub-lic and private companies at both the corporate and divisionallevels in line and staff assignments. During that time he heldkey executive positions with Herman Miller, RTE Corpora-tion, Union Special Corporation, and MGD Graphic Systems.He currently serves on the boards of directors of Harley-Davidson, Johnson Controls, Snap-on Incorporated, andBering Truck Corporation. Mr. Teerlink is a frequent speakerto business organizations internationally and an active par-ticipant in many business and community organizations.


HBR029ABO 8/15/01 1:39 PM Page 171

accountabilitymental disciplines and,

82–83PerkinElmer and, 93

acquisitions, at PerkinElmer,94, 95–97

After Action Review (AAR; U.S.Army), 73

disciplines in, 78–86review process and, 75–77

AlliedSignal, 89, 91, 93AMF, 135–136. See also

Harley-DavidsonAnalytic Instruments, 96Anders, William A., 4Andersen Consulting, 145,

146–147Archuleta, Phil, 33, 34ASDA, 2, 14–20AT&T, 119

balanced scorecard approach,119–124, 126–127, 131. Seealso Duke Children’s Hos-pital

Beals, Vaughn, 136best practices, 122–123

big, hairy, audacious goals(BHAGs), 25, 49–51

Body Shop, 31Bossidy, Larry, 101Brennan, Ed, 61–65, 63–64, 71Briggs, Roger, 40bureaucracy, and catalytic

mechanisms, 28, 29–32, 34

Campbell, John, 137–138Canavino, Jim, 152catalytic mechanisms, 23–54

characteristics of, 27–41defined, 25dreams and, 48–49evolution of, 46–47examples of, 28, 29–32,

33–37, 38–39individuals and, 51–54principles for building,

41–48traditional controls and,

27–41value of, 25

Champion International, 2,5–12

Citicorp, 50

Index

173

HBR029index 8/15/01 1:40 PM Page 173

net30

net30

coalitions, and grassrootschange, 150–157, 163

Coldwell Banker, 62combined strategies

catalytic mechanisms and,47–48

theories of change and, 3–4,12–20

“come to Boulder rule,” 44, 51communication

with customers, 136employee feedback and, 128,

131goals and values and, 100listening and, 118–119, 128,

131straight talk and, 80, 107,

109–111, 113strategic understanding and,

78–80survival strategies and,

130–131translators and, 164

Compaq, 145, 146conflict

healthy confrontation and,105–106, 109–111

revitalization and, 60, 64–65,68–70

consultants. See also Ozley,Lee

Harley-Davidson and, 138theories of change and, 6,

10–12, 19–20controls versus catalytic

mechanisms, 27–41

174 Index

co-opting, and grassrootschange, 164

Cope, Lance, 43cost, catalytic mechanisms

and, 45–46Cox, Bryan, 113creativity, and catalytic mecha-

nisms, 43–45credibility, 107–108culture

PerkinElmer and, 93–94of success, 111–112transformation of, and

CEOs, 133–143vital signs and, 59–60,

62–65culture-based change

combined with shareholdervalue approach, 12–20

compared with shareholdervalue approach, 5–12

entrepreneurial change and,21

general features of, 4limitations of, 12

DCH. See Duke Children’s Hos-pital (DCH)

Dean Witter, 62Deep Blue supercomputer, 157Digital Equipment, 145, 146,

154direction, sense of, 84Discover Card, 62dismissal, and employee incor-

poration, 68

HBR029index 8/15/01 1:40 PM Page 174

Fortune 100 companies, 86Fortune 500 companies, 119“four-day rule,” 44–45, 51Fujitsu, 145, 146fundamental change, factors

in. See revitalization

gain-sharing program, 134Gates, Bill, 156Gelb, Tom, 137–138General Dynamics, 4General Electric, 13–14, 89, 91,

93Gerstner, Lou, 147, 153, 156goals. See also vision, imple-

mentation ofachievability of, 106,

111–112, 131, 164–165balanced scorecard approach

and, 119–121, 131BHAGs and, 25, 49–51catalytic mechanisms and,

23–54future orientation and,

80–81for performance, 90, 93, 131revitalization leadership

and, 71–72sense of mission and,

117–118strategy and, 93theories of change and, 6, 7,

15Goldstein, Jeff, 43Goodbye Darkness (Man-

chester), 45–46

Index 175

divestment, at PerkinElmer, 94,97–98

dreams, and catalytic mecha-nisms, 48–49, 51–53

Duke Children’s Hospital(DCH), 115–132

balanced scorecardapproach at, 119–124,126–127, 131

cost versus mission and,118–119

employee involvement and,121–124

measurement system at,124–127

Dunlap, Al, 7, 8, 11, 12. See alsoScott Paper

EDS, 145, 146EG&G. See PerkinElmeremployability, 84employees. See human

resourcesemployment contract, 83–85entrepreneurial companies,

and change theories,20–21

failure. See setbacks, learningfrom

“15% rule,” 46–47focus

management discussionsand, 127

theories of change and, 6,8–9, 17

HBR029index 8/15/01 1:40 PM Page 175

Gore, Bill, 38–39Granite Rock, 25–27, 30, 32, 43,

47–48grassroots-based change

at IBM, 147–162steps in, 162–165

Grossman, Andrew, 153Grossman, Dave, 145–146,

147–152, 156, 157,158–162, 164

Harley-Davidson, 133–143authoritarian leadership at,

133–137Harley Owners Group (HOG)

and, 136inclusive leadership and,

140–143joint visioning process and,

137–140Hewlett-Packard, 4, 27, 42–43,

153honesty, 80, 107, 130Hornung, Lane, 45hospital turnaround, 115–132.

See also Duke Children’sHospital

human resources. See alsotraining programs

avoidance of top-downapproach and, 134–135

catalytic mechanisms and,28, 37–39

employee incorporation and,65–70, 74

individual approach and,108–109

176 Index

PerkinElmer and, 92–93,98–99, 100, 101

professional developmentand, 93, 99, 112

recruiting and, 98–99,112–113

IBMfirst Web demo at, 149–152“Get Connected” paper,

151–152grassroots Internet coalition

at, 150–157home page mock-up demo

at, 152–154Internet World exhibit,

154–155, 156missed opportunity and,

147–149official Olympic website by,

157, 158, 159public Internet development

process at, 158–162software development prin-

ciples and, 159identity, and revitalization, 60,

63–64incentives. See reward systeminformation systems, 124–127,

129Intel, 119, 145, 147Internet World (annual trade

convention), 154–155, 156investors, and PerkinElmer,

101–102. See also share-holder value approach

Isolab, 96

HBR029index 8/15/01 1:40 PM Page 176

net30

Maslow, Abraham, 139McKinsey & Company, 89, 91meaning, sense of, 84measurement systems,

124–127mental disciplines, 75–86

future orientation and,80–81

inventive accountability and,82–83

learning from setbacks and,82

quid pro quo and, 83–85straight talk and, 80strategic understanding and,

78–80Microsoft, 145, 147morale

feedback and, 128–129performance evaluation and,

127–128motivation, individual

approach to, 105, 107–108multidisciplinary team,

122–124

National Training Command(NTC; U.S. Army), 74–77

NEN Life Sciences, 97neutralization, and grassroots

change, 164New England Patriots, 107New York Giants, 107–108,

112New York Jets, 107Next (computer), 21Nike, 50

Index 177

joint vision-building process,139–140

Kaplan, Robert, 119Kasparov, Garry, 157Kimberly-Clark, 41Knight, Chris, 66–70, 81Kohnstamm, Abby, 149, 150

leadershipauthoritarian, 136–137cultural transformation and,

133–143employee involvement and,

133, 138–140at Harley-Davidson,

133–143inclusive approach and,

140–143at PerkinElmer, 100revitalization approach to,

70–75taking charge and, 107–108talent development and, 93,

99, 112theories of change and, 6,

7–8, 16–17Leighton, Allan, 16–17, 19–20Life Science Resources, 96–97listening, 118–119, 128, 131

Malcolm Baldrige NationalQuality Award, 27

Manchester, William, 45manifesto, and grassroots

change, 163Martinez, Arthur, 62, 71–72, 80

HBR029index 8/15/01 1:40 PM Page 177

nonprofit organizations. SeeDuke Children’s Hospital

Nordstrom, 25, 83Norman, Archie, 15, 16–17,

18–20. See also ASDANorton, David, 119Nucor Corporation, 35–37, 38,

46

observer/controllers (U.S.Army), 73

OlympicsSummer (1996), 157Winter (1994), 147–149

ongoing effects, and catalyticmechanisms, 28, 39–41,49–51

operating plan, 93organizational drift, 60–65organizational learning, 60. See

also training programsOzley, Lee, 138–140, 142, 143

Pagonis, Gus, 82patient care protocols, 122–123Patrick, John, 146, 150–162, 164PE Corporation, 96performance evaluation, 100,

109information systems and,

124–127self-evaluation and, 127–128

performance indicators. Seevital signs

PerkinElmer, 89–104human resources and,

92–93, 98–99, 100, 101

178 Index

name change from EG&G,93, 101–103

training initiatives at,103–104

Petronas (company), 67, 81Philip Morris (company), 50point of view (POV), 162–163POV. See point of view (POV)power distribution. See also

grassroots-based change;leadership

avoidance of top-downapproach and, 134–135


revitalization and, 60, 63process, and theories of

change, 6, 9–10, 17–18programs. See also revitaliza-

tionelimination of, 42–43fundamental change and, 55,

56–57, 85–86theories of change and, 6,

9–10, 17–18public perceptions

company name and,101–102

Internet development atIBM and, 158–162

quid pro quo, levels of the,83–85

recruiting, 98–99, 100red flag device, 30–32, 46resocialization

HBR029index 8/15/01 1:40 PM Page 178

understanding the quid proquo and, 83–85

Roddick, Anita, 31Rotron, 94Royal Dutch Shell, 56, 58–60,

65, 83–86. See also ShellMalaysia

SAS. See StrategicVision Soft-ware (SAS)

Scott Paper, 2, 5–12Sealol, 94Sears, Roebuck & Company, 56,

58–60, 61–65, 71, 80, 82,83–86

sequencing approaches tochange, 13–14

setbacks, learning from, 82,132, 140

shareholder value approachcombined with cultural

approach, 12–20compared to cultural

approach, 5–12entrepreneurial change and,

20–21general features of, 3–4limitations of, 11–12sequencing and, 13–14

Shell Malaysia, 66–70, 80, 81.See also Royal Dutch Shell

“short-pay” policy, 26–27, 30Sigler, Andrew, 7, 9. See also

Champion InternationalSinger, David, 151Smith, Darwin, 41Sony, 50–51

Index 179

leading from a differentplace and, 72–75

mental disciplines and, 75–86“Respond to Unreasonable

Customer Requests”(Nordstrom motto), 83

restructuring, at PerkinElmer,93–95

resultsachievable goals and, 106,

111–112, 131, 164–165catalytic mechanisms and,

28, 35–37continuous improvement

and, 90credibility and, 107–108customer satisfaction and,

117, 130discomfort with the status

quo and, 85–86IBM and, 164–165measurement system and,

124–127revitalization, 55–87

employee incorporation and,65–70

leadership and, 70–75mental disciplines and, 75–86organizational drift and,

60–65performance indicators and,

59–60sources of, 56, 58

reward systemPerkinElmer and, 93, 99theories of change and, 6, 10,

18–19

HBR029index 8/15/01 1:40 PM Page 179

net30

sports teams, 105–113status quo, creation of discom-

fort with, 85–86StrategicVision Software (SAS),

125Sullivan, Gordon R., 73Sun Microsystems, 145–146,

148–149, 153

targets, and grassroots change,163–164

team off-site meetings,140–142

Texas schools, 40–41theories of change, 1–21

basic archetypes, 1–2, 3combined strategies and,

3–4, 12–20comparisons between, 5–12limitations of pure forms of,

11–12reconciliation strategy and,

14–20sequencing and, 13–14soft versus hard approaches

and, 4–5Thoreau, Henry David, 493M, 29–30, 46–47, 119training programs

PerkinElmer and, 93, 99,103–104

U.S. Army and, 72–75translation, 164

U.S. Army, 56, 58–60, 65, 83–85training centers, 72–75

U.S. Constitution, 33

180 Index

U.S. Marine Corps, 33–34, 45,46

U.S. Open Web site, 157UNIX, 148, 149urgency, sense of, 71–72, 100USAA, 85–86

valentines exercise, 69–70values

catalytic mechanisms and,50–51

communication and, 100sense of mission and,

117–118vision, implementation of,

139–140vital signs

organizational drift and,62–65

revitalization and, 59–60Vivid Technologies, 96

W. L. Gore & Associates, 38–39waiver-of-regulation requests,

34–35Wallace, William S., 77–78Welch, Jack, 13–14Wladawsky-Berger, Irving, 150,

158Wood, Robert, 63, 81Woolpert, Bruce, 25–27Woolpert, Steve, 25–27work environment


employee involvement and,133, 138–140

HBR029index 8/15/01 1:40 PM Page 180