moves that matter: issue selling and...

® Academy of Management Journal2001, Vol. 44, No. 4, 716-736.

MOVES THAT MATTER: ISSUE SELLING ANDORGANIZATIONAL CHANGE

JANE E. DUTTONSUSAN J. ASHFORD

University of Michigan

REGINA M. O'NEILLSuffolk University

KATHERINE A. LAWRENGEUniversity of Michigan

In this study, we examined 82 accounts of "issue selling" to better understand man-agers' implicit theories for successfully shaping change from below by directing theattention of top management. The study reveals the importance of various issue-sellingmoves, including packaging, involvement, and timing. Managerial accounts uncoverthree kinds of contextual knowledge that are critical to the execution of issue-sellingmoves. The study shows managers actively shaping the issue-selling microprocessesthat contribute to organizational change.

Scholars often porfray change as a discrete event—a system is imfrozen, moved, and refrozen (Levmi,1951). But in reality, organizations are a cacophonyof complementary and competing change attempts,with managers at all levels joining the fray andpushing for issues of particular importance tothemselves. Indeed, it may be most accurate toportray an organization as a pluralistic marketplaceof ideas in which issues are "sold" via the persua-sive efforts of managers and "bought" by top man-agers who set the firm's strategic direction. Effec-tive management in this second portrayal of changeis of critical importance to the increasing number offirms that must compete in high-velocity environ-ments through a process of continuous change(Brown & Eisenhardt, 1997). Effective managementbegins with an understanding of a key componentprocess, issue selling.

Issue selling is the process by which individualsaffect others' attention to and understanding of the

We thank Brian Golden, Karen Golden-Biddle, KarenGramalla, Stuart Hart, Jane Howell, Karen Locke, DebraMeyerson, Leslie Perlow, Anjali Sastry, Scott Serich,Frances Westley, and Karl Weick for their helpful com-ments on drafts. We also appreciate the assistance of PhilBobko on an earlier version as well as the comments ofHerminia Ibarra and our anonymous reviewers. Thestudy benefited from the support of the ManagementSimulations Project at New York University and the Wil-liam Russell Kelly and Michael and Susan JandernoaChairs of Business Administration at the University ofMichigan.

events, developments, and trends that have impli-cations for organizational performance (Ansoff,1980; Dutton & Ashford, 1993). Because no issue isinherently important or strategic, individuals'claims about what matters (that is, their issue sell-ing) determine, in part, which change initiatives getactivated (Dutton & Ashford, 1993; Floyd & Wool-dridge, 1996). Issue selling shapes an organiza-tion's investment of time and attention and therebyshapes, in part, the actions and changes that ensue.Issue sellers are "players" (Ocasio, 1997) who use arepertoire of moves to sell issues and affect top-level decision makers' attention. By moves, wemean the behaviors that constitute an interaction(Goffrnan, 1981; Pentland, 1992). In this study, wegained insight into managers' implicit theories ofissue selling—their sense of what works anddoesn't work in this process—by asking them todescribe issue-selling episodes. From a qualitative,systematic examination of tbese descriptions, weextracted lessons about the elements thought to beimportant to successful issue selling. Before turningto our study, we first show how a focus on issueselling can provide unique insights into more generalchange processes. This study also addresses a gapthat we identify in the current literature on issueselling. Finally, we clarify the assinnptions uponwhich we based our data gathering and analysis.

ISSUE SELLING AS A GHANGE PROGESS

Issue selling is an early part of a more generalchange process. Issue selling affects the allocation

716

2001 Dutton, Ashford, O'Neill, and Lawrence 717

of management attention, a limited and scarce re-source in organizations (March & Shapira, 1982;Pfeffer, 1992; Simon, 1947). As depicted in atten-tion-based views of the firm (March & Olsen, 1976;Ocasio, 1997), patterns of attention allocation in-side a firm partially explain patterns of organiza-tional change. What is important is how attentionallocation activates some decisions and not others(Dutton, 1997; Kingdon, 1990; Langley, Mintzberg,Pitcher, Posada, & Saint-Macary, 1995; Ocasio,1997; Weiss, 1989). An issue-selling perspectivedirects attention to the ofren-unnoticed acts ofchange agents, below or outside organizations' topmanagement groups, who invite consideration ofsome issues and not others. This perspectiveframes people outside the top management team aspotentially potent initiators of change (e.g., Floyd &Wooldridge, 2000; Nonaka, 1994; Stumpf &Mullen, 1992). It helps to rewrite change as a moreemergent and pluralistic process than it is typicallyassumed to be in more traditional "upper echelons"perspectives (e.g., Finkelstein & Hambrick, 1996;Hambrick, 1989; Hambrick & Mason, 1984).

Furthermore, an issue-selling perspective allowsrecognition of the inherent intersection of microand macro forces in determining change patterns inorganizations. It highlights the social-psychologicalunderpinnings of change processes by supportingtheorizing about individuals' motivation to under-take efforts to shape the attention of others butimplies that such activities are always contextuallysituated (Gollins, 1981; Ocasio, 1997). Issue sellersact in efficacious and self-motivated ways to makechange happen (Scully & Greed, 1998), but within astructured context composed of rules, routines, val-ues, and norms that constrain what they do. Wepropose that, over time, they develop implicit the-ories, constituting theories-in-use (Argyris &Schon, 1982), based on knowledge about the con-text. Their theories inform how they attempt to gainattention for their issues and make change happen.

Past Research on Issue Selling

Our in-depth study of issue-selling moves alsocontributes to the more specific research on issueselling. First, the study grounds a conceptualmodel of issue selling proposed by Dutton andAshford (1993) with accounts of actual issue-sell-ing episodes as they occured in an organizationalsetting. To date, the empirical work on issue sellinghas been limited to two types. First, there are stud-ies of the factors that contribute to issue sellers'decisions to sell an issue. For example, Dutton,Ashford, Wierba, O'Neill, and Hayes (1997) exam-ined the contextual cues issue sellers used in as-

sessing whether or not to undertake issue selling.Ashford, Rothbard, Piderit, and Dutton (1998) as-sessed the social-psychological processes throughwhich context affects an issue seller's willingnessto sell issues. Other studies have examined theclaims individuals make about the needs of partic-ular identity groups (Greed & Scully, 2000; Har-quail, 1996; Meyerson & Scully, 1995) as a type oforganizational issue. These latter studies started touncover the role of context in shaping how claimsare made about what issues are important. How-ever, the focus of these studies has been on claimsnarrowly related to identity issues; the issues in thecurrent study are much broader in range.

The second type of empirical effort establishes arelationship between middle managers' involve-ment in processes such as issue selling and organi-zational performance (Floyd & Wooldridge, 1996;Westley, 1990; Wooldridge & Floyd, 1990). How-ever, little empirical evidence exists about thekinds of moves that compose the actual issue-sell-ing process. Researchers have studied the anteced-ents and processes used by innovation champions(Howell & Higgins, 1990), by whistle-blowers(Miceli & Near, 1992), and by change initiators(Kanter, 1983; Morrison & Phelps, 1999) who en-gage in related forms of discretionary action, butunderstanding of the way that people accomplishthis process remains incomplete.

On the basis of tbeory but no data, Dutton andAshford (1993) used impression management andupward influence research from psychology andsocial problem theory from sociology to build aninitial framework for thinking about issue sellingand issue-selling moves. They identified movesfrom four general categories. The first category in-cludes packaging moves, such as the content fram-ing of an issue (for example, implying responsibil-ity), presentation of the issue (for example, inemotional or novel terms), the type of appeal used(for example, one- or two-sided), and the bundlingof the issue with other issues to enhance the chancesof gaining attention. The second, third, and fourthcategories of moves—^which they group under thelabel "process" but we separate here—include in-volvement moves, choice-of-channel moves, and for-mality moves. Our goals for the present study were tosee if these two types of moves showed up in actualmanagers' descriptions of the process and to allowthe data to suggest new categories and add nuance orcomplexity to the prior descriptions.

Assumptions

We assumed that managers' accounts of themoves they made in successful and unsuccessful

718 Academy of Management Journal August

issue-selling episodes could yield a better under-standing of the process of issue selling. Moves cre-ate new situations from existing situations by ac-complishing desired ends (Pentland, 1992). Goffman(1981) proposed the move as a unit of analysis ininteractions. Although the relationship between man-agerial accounts of their issue-selling moves and theiractual selling process is imperfect and fraught withthe usual biases of recall (Golden, 1992; Schwenk,1985), we treat these accounts as imperfect represen-tations of managers' implicit theories evoked to ex-plain unsuccessful and successful selling moves.They are a frace of managers' "sense making" aboutwhat works and does not work for selling issues intheir organizational setting (Weick, 1995).

Our reliance on managers' stories as useful datafits with research on how people understand (storeand remember) their own experience and with therecent wave of interest in narrative analyses of or-ganizational phenomena (e.g., Barry & Elmes, 1997;Boje, 1991; Gzarniawska, 1997; Martin, Feldman,Hatch, & Sitkin, 1983; O'Gonnor, 1997,1999; Weick& Browning, 1986). In the words of Bruner, narra-tives are a "natural vehicle for folk psychology.[They deal] 'with the stuff of human action andhuman intentionality" (1990; 52). Stories provide a"wedge into the expertise" of issue sellers (Klein,1998; 193).

Our second assumption was that the moves de-scribed by issue sellers help to explain a piece ofthe "knowledgeable bebavior of people tbat ac-counts for all organizational activity" (Floyd &Wooldridge, 2000; 131). Moves are actions that areinformed by and expressive of practical knowledge(Pentland, 1992). In this case, the knowledge isabout how to have influence and make change hap-pen in organizational contexts. Such loiowledgemay be a form of practical intelligence that manag-ers acquire with experience (Wagner & Sternberg,1985). It may also be a form of social expertise thatis acquired and revised through interactions withothers (Fligstein, 1997). The current study appliesinsights from practice theory (Bourdieu, 1977;Lave, 1988) and practice theory's emphasis on sit-uated Icnowledge to better explain the practical ac-complishment of directing attention to issues inorganizations.

Although we began with the assumption thatissue-selling moves are a form of intelligent indi-vidual action, we did not originally intend to focuson the knowledge that informed moves as a keytheme. However, interviewees' accounts providedsubstantial evidence about the knowledge neededand used for issue-selling moves. Given this evi-dence, we analyzed our data to better understandthe knowledge that enabled specific moves.

METHODS

Organizational Setting

The setting for this data collection was a large(411 beds, 2,683 employees), not-for-profit regionalhospital located in a rural northeastern town andconnected to a prestigious educational institution.The hospital, which we labeled "Northeast Hospi-tal" for this study, is one of three entities (alongwith an outpatient clinic and a medical school)composing a medical center that is a regional pro-vider of medical services to a large geographicalarea in the Northeast. We focused on issue sellingwithin the hospital entity but note that the hospi-tal's connections with the clinic and medicalschool were an important contextual feature.

It is helpful to understand the nature of the hos-pital as a context for issue selling. Northeast Hos-pital was the dominant hospital (and the only tertia-ry-care facility) in the region. Throughout much of itsexistence. Northeast Hospital had enjoyed near-mo-nopoly status, and its strategy was conservative, risk-averse, and internally rather than externally (that is,customer) focused. At the time of this study, how-ever. Northeast Hospital's decision makers con-fronted many of the same competitive pressures thatchallenge most hospitals today, including a need tolower costs, focus on quality, and accommodatethird-party payers (Shortell, Morrison, & Friedman,1990). These challenges translated into a heightenedinterest in, and pressure for, innovation within thehospital. Also, in its recent history (within the sixmonths prior to the study), the hospital had experi-enced a severe budget crunch that mobilized forces atthe senior management level to respond to a dovra-turn in volvmie and revenue losses. Managers oftenreferred to this budget crunch and to the constructionof a new facility to which the entire center wouldrelocate the following year.

Sample and Procedures

The hospital structure had three managerial lev-els. The top management team consisted of fiveindividuals; the president, and the senior vice pres-idents of planning and marketing, operations, fi-nance, and nursing. The next level in the hierarchyconsisted of 15 vice presidents, each reporting toone of the four functional senior vice presidents. Atthe third level, 62 department directors each re-ported to one of the 15 vice presidents. The samplein this study consisted of all the vice presidentsavailable during the time of the study (13 out of 15)and a randomly selected subsample of the depart-ment directors (29 out of 62). This final sample of42 middle-level managers was distributed over the


four major functional areas in the hospital—nurs-ing [n = 12), finance (n = 3), operations (n = 23),and planning and marketing [n = 1)—and includedothers who reported directly to the president (n =3). The sample distribution over functional areasmirrored the relative sizes of the four departmentsin the hospital. Members of the sample held anaverage of 8.4 years of tenure within the hospitaland had an average tenure of 4.8 years in theircurrent positions. Fifry-two percent of them weremen.

The vice president of operations, our main con-tact person, sent a memo to all vice presidents anddepartment directors explaining the study and thehospital's decision to participate in it. The memoinformed employees about the random process forselecting interviewees and told them that theymight be contacted for an interview. We tben con-tacted the random sample of employees to arrangeinterviews. The interview protocol had several sec-tions. First, we described the goals and nature ofthe project and asked interviewees to provide gen-eral information about their backgrounds (for ex-ample, education, tenure in hospital), their currentpositions in the organization (for example, title,number of "direct reports"), and tbeir extent ofdirect and indirect contact with the top manage-ment group. Next we asked them to think aboutparticular incidents in which they had tried to di-rect the attention of the top management to a par-ticular issue. They were asked to define an issuevery broadly—as any event, development, or trendthat could affect the hospital's performance; thisdefinition of a strategic issue is from Ansoff (1980).

We then asked the interviewees to describe, in anopen-ended fashion, a successful attempt on theirparts to bring an issue to the attention of top man-agement. We asked them to describe the majormilestones in the process and how they interpretedthem. Our goal was to capture, as completely aspossible, a retrospective account of the process as itunfolded over time and the elements that seemedmost important in the interviewees' understandingof the story. Then this protocol was repeated whenwe asked respondents to describe an unsuccessfulattempt to sell an issue to top management. Thus,our data collection methods were open-ended inthe sense that the interviewees could emphasizesome aspects of issue selling over others and raiseany aspects they thought made a difference. Theprotocol was directive, too; while an intervieweedescribed a process, we probed specifically aboutwho was involved in the process, how they wereinvolved, what seemed to work well, and whatthings the seller could have done differently. Thefinal portion of the interview asked the managers to

describe the hospital (what made it unique anddistinctive), the top management group, and thetypes of broad issues considered important andunimportant in the hospital.

The unit of analysis for our results was an issueepisode. Our data consisted of a total of 82 issueepisodes generated by 42 interviewees. Forty inter-viewees described both successful and unsuccess-ful issue-selling episodes, and 2 people chose totalk only about a successful episode.

Goding

The goal of the coding scheme was to capture themajor moves that interviewees mentioned, in theirdescriptions of successful and unsuccessful issue-selling episodes. Following Miles and Huberman(1994), we developed the coding scheme induc-tively, adding new codes as the interviewees men-tioned new types of moves in the different inter-views. In this sense, our final set of codes wascomprehensive—it included a complete list of cat-egories that served as retrieval and organizing de-vices (Miles & Huberman, 1994; 57) relating to aparticular concept, theme, or idea that intervieweesmentioned in their issue-selling accounts. The cod-ing process involved three steps.

First, two of the authors began with a "start list"(Miles & Huberman, 1994) of codes based on themajor descriptive categories of behaviors that Dut-ton and Ashford (1993) used to describe the processof issue selling; packaging moves, involvementmoves, type-of-channel moves, and formalitymoves. We did not code the characteristics of thetop management team or issue seller characteristics(which were part of the Dutton and Ashford mod-el), as they did not capture the process involved inissue selling.

Second, the same two coders independently readthe transcripts of five interviews in order to de-velop an exhaustive list of the codes that emergedfrom these interviews. The two coders then createda more complete start list of codes that captured themajor categories of moves and subcategories thatthese respondents used in their issue-selling de-scriptions. The research team created a brief de-scriptive label for each category and subcategorythat allowed the coders to assess whether the movewas mentioned in each issue-selling episode. Forexample, within the general category of issue pack-aging, a subcategory emerged tbat was labeled "usethe logic of the business plan in presentation." Anepisode was coded as suggesting this subcategorywhen interviewees mentioned that the issue wasframed in a way that used a business plan logicand/or process. Each successful and unsuccessful


issue for each interviewee was coded as a separateepisode.

Interviewees who described their selling effortsin ways consistent with a subcategory in an episodewere given a code of 1 for that subcategory, andinterviewees who did not mention that subcategoryin their episode received a 0. In this tally, we in-cluded both explicit and implicit mentions. In anexplicit mention, the interviewee used the words ofthe subcategory title or definition, and in an im-plicit mention, the interviewee's comment cap-tured the essence of the subcategory without usingthe exact words.

Third, we only included a category in our find-ings if it was mentioned in at least eight (10%) ofour sample of episodes. This procedure led to sub-categories. The Appendix presents the final list ofcategories and their definitions. This list includesfive broad categories that describe issue selling (forexample, packaging, involvement, formality, prep-aration, and timing) as well as various subcatego-ries that describe each theme at a finer level ofdetail. The total category set captures most of thecontent of the interviewees' accounts of their suc-cessful and unsuccessful issue-selling episodes.Each coder completed the full coding task. Follow-ing Miles and Huberman's (1994) formula for de-termining agreement, we had 60 percent of thecoded material check-coded by another coder, ob-taining approximately 85 percent agreement.Where disagreements occurred, the coders dis-cussed their options until they reached agreement.

Analyses

We performed two analyses on the coded inter-view data and then did a third analysis of thecontextual knowledge that enabled the moves. Thefirst involved a simple count, for each category, ofthe number of interviewees who mentioned movesin that category (either explicitly or implicitly). Weused this count to gauge the salience of moves inissue sellers' overall thinking. If a particular personmentioned a move more than once in an episode,we still counted it as a single mention. We consid-ered counting the absolute number of mentions, butsuch a count might have reflected individual dif-ferences in verbal style and overweighted the im-portance that particular issue sellers placed on thedifferent moves. This conservative counting rulemeant that the total number of mentions serves as arough indicator of the relative salience of the vari-ous categories of issue-selling moves in people'simplicit theories of issue selling.

The second analysis compared counts of men-tions of moves in both successful [n = 42) and

unsuccessful (n = 40) selling episodes. To do this,we counted separately the number of times an in-terviewee said he or she either did use or did notuse each move during both successful and unsuc-cessful issue-selling episodes. Each time interview-ees stated that they did not use a move, we alsocoded whether or not they mentioned that, in ret-rospect, they wished they had used the move.Phrases that indicated such regret included; "I wishI could have done," "I tried to do it but I couldn't,"and "I could have done a better job if I had." Weinterpreted these types of statements as indicationsthat an interviewee believed the move would havemade the selling effort more successful. We thencompared the extent to which moves were dis-cussed as used or not used by managers in order togain further insights into their relative importancein their theories of successful and unsuccessfulissue selling.

We also undertook a third analysis we did notanticipate when the study began. In coding thedata, we noted that interviewees frequently re-ferred to various kinds of knowledge about featuresof the organizational setting (which we call "con-textual knowledge") that encouraged, enabled, orhampered them in executing a move. Our thirdanalysis involved a additional pass through all ofthe episode data to document the kinds of knowl-edge about Northeast Hospital the sellers describedas informing their capacity to execute issue-sellingmoves. We examined the moves themselves to seewhat knowledge about the context they indicatedor implied was necessary. We also noted what fea-tures of the context were identified as pathways orroadblocks to success. By "pathways" to success,we mean aspects of an issue-selling context that aninterviewee indicated were helpful and thus gavethe issue a greater likelihood for success. By "road-blocks," we mean aspects of a context tbat a respon-dent indicated were troublesome, a hindrance, orlikely to have caused the failure of the issue-sellingattempt. Although pathways and roadblocks madethe context more or less favorable for selling anissue, knowledgeable navigation of these contex-tual factors facilitated an issue seller's efforts, in-dependent of prior conditions. From this reassess-ment of interviewees' descriptions of moves,pathways, and roadblocks, we were able to derivethree types of contextual knowledge that were con-sidered by issue sellers as they selected moves.

The Issues

The issues that managers "chose were varied.Some were highly specific, tangible solutions to aperceived problem (for example, the need for an


in-house mini-laundry). Others were more nebu-lous issues judged as worthy of organizational at-,tention but perhaps lacking immediate or imaginedsolutions (for example, the need for some nursingdepartments to update old-style practices). Morethan half of the issues had a financial element(potential cost savings, revenue enhancements, orcapital outlays), such as the controversial billingfor the cost of drawing blood samples or the requestfor an increase in the dialysis budget. The issuesalso frequently addressed administrative, logistic,or technical functions at the hospital, includingrecord-keeping systems, scheduling processes, andtechnological aids. For example, one long-time em-ployee sold the administrative and logistical issuethat the hospital units should consolidate their reg-istration system. Many initiatives were patient- oremployee-related, involving the treatment, educa-tion, or organization of these constituencies, in-cluding issues around scheduling, safety, conve-nience, and information exchange. For example,one issue that impacted both employees and pa-tients was the proposal for a method of enhancingpatient and staff safety when a patient became vi-olent. Few of the issues were mentioned by morethan one manager. Although some of the issues weheard described must bave competed for attentionsimultaneously, no sense of competition concretelyemerged from the issue descriptions or selling ac-counts—interviewees did not mention the otherissues being sold at the same time they were sellingtheir own issues.

PATTERNS IN AGGOUNTS OF ISSUE SELLING

Below, we discuss the patterns in interviewees'descriptions of issue selling in two sections. First,we present the categories of moves mentioned inthe issue-selling stories and compare them to thosedescribed in Dutton and Ashford's (1993) originaldescription of issue-selling actions. The three gen-eral categories of moves are packaging, involve-ment, and process-related moves; within each cate-gory, we discuss related types of moves. Then wecompare accounts of perceived successes and fail-ures in issue selling to discern what managers be-lieved worked or did not work. We view differ-ences between the mentions of moves made insuccessful and unsuccessful selling as imperfecttraces of what actually works and does not work inthis setting. Thus, these accounts are botb theoriesof effective and ineffective practices in this settingat the same time that they are retrospective attribu-tions about successful and unsuccessful actions.Second, we develop and elaborate a three-categoryframework of contextual knowledge that became

apparent as we analyzed the moves. We elaboratedthis framework by revisiting our data to identify thedifferent kinds of contextual knowledge implied inthe accounts of successful and unsuccessful issueselling. Both parts develop a theory of knowledge-in-use in issue selling and, correspondingly, inmaking change happen from below.

Interviewees raised many of the original processmoves proposed by Dutton and Ashford (1993)(packaging, involvement, channel type, and formal-ity); several new process elements also appeared inthe accounts. In addition, the issue episodes revealelaborate sensibilities about various packaging andinvolvement cboices tbat were not anticipated inthe earlier issue-selling framework. We use thesenew process moves and a deeper exploration of the"old" process moves to develop a knowledge-basedaccount of how managers think about issue-sellingprocesses in organizations.

Packaging Moves

On the basis of Dutton and Ashford's (1993) the-oretical work, we divided the packaging moves intotwo groups; presentation moves, which were as-pects of how an issue seller went about activelypromoting an issue, and bundling moves, whichwere sellers' attempts to connect the issue to otherissues or goals. Table 1 gives the frequencies of ourpackaging categories and subcategories.

Presentation. We labeled the most frequentlymentioned presentation tactic "using the logic of abusiness plan." Interviewees empbasized the im-portance of using lots of numbers and charts, con-veying a logical and coherent structure, and em-phasizing bottom-line impacts. Sometimes sellersdescribed concentrating on financial aspects (forexample, "running the numbers"), and at othertimes they focused on business perspectives moreclosely aligned with strategy, operations, or mar-keting. This move is consistent with Dutton andAshford's (1993) assertion that supporting facts andevidence will be important in tbe claim for anissue's legitimacy. One issue seller, who was tryingto make the case for more critical care space in theintensive care unit (IGU), illustrated her use of thelogic of a business plan;

So, it involved a lot of work with physicians, a lot ofwork with our data people, looking at trends andvolumes, and our current volumes and all that . . .So it involved the development of a full businessplan from a financial standpoint, a marketing stand-point, and a demographic assessment standpoint.

More than 75 percent of the interviewees men-tioned the use of a business plan logic for present-


TABLE 1Packaging: Presentation and Bundling

Category and Subcategory

PresentationUse logic of business planMake continuous proposals'^Package issue as incrementaP

BundlingTie issue to valued

goals—Profitability'^Tie issue to valued goals—Market

share/organizational image'̂Tie issue to concerns of

key constituents'^Tie issue to other issues'^

Total Mentionsby Respondent"

322310

21

17

23

22

Total

392712

31

23

30

27

Did Use

2710

8

18

15

15

17

Mentions by Episode''

Successful

Wisb HadUsed

000

0

0

0

0

Did NotUse

070

0

0

0

0

Did Use

981

2

'2

13

8

Unsuccessful

Wish HadUsed

112

6

3

2

1

Did NotUse

211

5

3

0

1

" Mentions are combined for each interviewee's successful and unsuccessful issues [n = 42 individuals]. If an interviewee mentioneda move more than once, it was counted as a single mention.

^ Mentions are separated by successful and unsuccessful episodes [n = 82 episodes) and tallied according to the framing of the mention.If an interviewee did not explicitly or implicitly mention a move as something he or she either used, did not use, or wished he or she hadused, then it was not mentioned at all.

'̂ Not proposed by Dutton and Ashford (1993).

ing an issue. This finding is consistent with influ-ence research data showing that rationality is verypervasive in upward influence attempts (Kipnis &Schmidt, 1983). A seller's actions may also createlegitimacy by couching an issue in a form that isconsistent with the language of the organization(Schriesheim & Hinkin, 1990). Gomparing the pat-tern of mentions strengthens this i.nterpretation, asthis move was described in many more successfulepisodes than unsuccessful episodes.

The second most frequently mentioned presenta-tion tactic was continuous proposal making. Morethan half of the interviewees mentioned this fea-ture, which involved raising issues many timesover a period of time. The goal in many cases ap-peared to be preparing tbe target to better "hear" afull proposal. For example, an issue seller de-scribed bis efforts to sell the issue of improvingcommunity service to a senior vice president in thehospital; "When you tell him about a concept, yousort of acclimate him to the situation and you re-peatedly tell him about it for several months so heknows it is coming, and he knows what is happen-ing. And then you hit him with the big package."Although continuous proposal making was fre-quently mentioned by sellers, there is no meaning-ful pattern in the distribution of mentions acrosssuccessful and unsuccessful episodes. Thus, weconclude that this move is a part of managers' the-

ories of how to promote an issue at Northeast Hos-pital, but their opinions about its contributions tosuccessful issue selling remained mixed. It may bethat the reliance on continuous proposal making ismostly a function of the nature of managerial work(fragmented and discontinuous, with only shortperiods available for any activity) (Mintzberg,1997). As a result, sellers may rarely have thechance to present issues all at once, so they bringthem forward over time in bits and pieces.

Presenting an issue as if it were an incrementalchange was mentioned by 24 percent of the inter-viewees. Although this tactic seems similar to con-tinuous proposal making, it refers to the size of thechange implied by an issue, not to tbe multiple(possibly small) presentations used to prepare thetarget to hear the complete issue. That is, an incre-mental issue may be presented all at once, whereascontinuous proposal making serves to make a rad-ical or incremental issue more familiar through re-peated presentations. In their explanations for whyincremental presentations worked (or did not), is-sue sellers noted that making an issue seem incre-mental made it more palatable for issue targets.Such statements support Weick's (1984) idea that"chunking" a social problem into a series of smallwins may induce greater acceptance than trying toattack the full-scale problem. The pattern is alsoconsistent with Frohman's (1997) studies of per-


sonal initiators of change who tended to use incre-mental as opposed to revolutionary action patterns.An issue seller illustrated this logic in his accountof a failed effort; "I lumped a lot of different thingstogether, and I might have tried to do smaller pieces."In the successful episodes in our data, each timethis move was mentioned, the interviewee hadused it; but most of the mentions (three of four) inthe unsuccessful episodes were of not using it.Thus, sellers'^sense making about this move sug-gests agreement that this kind of presentation tacticis one that they believe works.

Issue bundling. The accounts revealed four waysthat issue sellers talked about connecting, or bun-dling, an issue with other ideas that had currencyin the organization. In their accounts, the bundlingmoves were much more varied and prominent thanDutton and Ashford (1993) proposed. Sellers de-scribed tying their issues to the organizational goalof profitability, to market-related issues, to the con-cerns of key constituents, or to other issues. Each ofthese bundling moves was mentioned by more than40 percent of the interviewees. In addition, inter-viewees only talked about using bundling in thesuccessful episodes, whereas they talked aboutboth using and not using bundling in the unsuc-cessful stories.

The most clear-cut case of a theory of "whatworks" in bundling is in the case of tying an issueto valued goals like profitability. For this issue-selling move, there were mentions of its use in 18successful accounts, whereas only 2 accounts ofunsuccessful episodes contained mentions of usingit. Six interviewees describing unsuccessful at-tempts said they would have used it in retrospect.The move of tying an issue to the valued goals ofmarket share and a positive organizational image(more externally focused goals) followed a compa-rable pattern. Thus, sellers tended to believe thatconnecting an issue to valued goals is associatedwith effective issue selling.

One informant illustrated the success of issuebundling in her attempt to sell the need for morecritical care facilities at Northeast Hospital. In thisquote, she clearly ties the issue to the valued organ-izational goal of enhancing patient care and im-proving revenues;

Well, there were several overriding concerns. One. . . was patient care. The issue with patient care wasthat we're not able to provide all the services weneeded to provide to all of our patients because wedidn't have the space . . . So, the first "couch" wasprobably patient care. The second "couch" waseconomy, I mean the finance of turning away pa-tients is pretty . . . it struck quite a chord shall we say.

Less clear from the bundling data is the utility ofthe moves aimed at tying an issue to concerns ofkey constituents and tying an issue to other issues.Although both of these issue-selling moves werementioned by more than half of the interviewees,they were used in both successful and unsuccessfulissue-selling episodes. Interviewees mentioned ty-ing their issues to concerns of key constituentsalmost as often when describing successful as un-successful accounts. Similarly, sellers mentionedtying an issue to other issues in accounts of bothsuccessful and unsuccessful issue selling, but twiceas many mentions were in successful episode de-scriptions. Thus, we conclude that these two bun-dling moves are part of managers' implicit tbeoriesof how to sell issues at Northeast Hospital, but theirassociation with success is mixed. The pattern ofbundling findings suggests tbat, in contrast to Dut-ton and Ashford's (1993) observations, it is thelinking of one's issue to already agreed-upon goalsrather than to other issues currently up for consid-eration that people believe is associated with issue-selling success.

Involvement Moves

Issue sellers in organizations face choices ofwhom to involve in their issue-selling efforts. In-novation research (Dean, 1987; Kanter, 1983), strat-egy process research (Burgelman & Sayles, 1986;Lyles & Reger, 1993; Wooldridge & Floyd, 1990),coalition-building studies (Dutton & Webster, 1988;Murnigban, 1978), and participation research(Wagner, 1994) all suggest that involving othersduring upward influence attempts can positivelyaffect the scope and the impact of one's efforts. Byinvolving others, sellers often intend to make an issuemore visible to yet other organization members, to"grow" a larger and potentially more powerful coali-tion, to build commitment to the issue and its reso-lution, and to ensiu-e that attention and action aredevoted to the issue. Our interviewees clearly sup-ported Dutton and Ashford's (1993) contention thatwhether to involve others or to "go solo" is an impor-tant selling consideration. The 1993 article is alsoquite explicit about the nature of others' involvement,specifically, its formality and the range of other peo-ple who can be involved. Although Dutton and Ash-ford never mentioned these involvement dimensions,the data suggest that these are key aspects of theissue-selling moves in this context.

Sellers talked about the targets of involvementand the nature of involvement (its formality andbreadtb). Table 2 gives the frequencies of our in-volvement categories and subcategories. Sellersalso discussed the timing of involvement, which


TABLE 2Involvement: Targets and Nature


Targets of involvementInvolve others at upper level'̂Involve others at same level'̂Involve others outside organization'^Involve others at lower levePKeep boss informed'̂

Nature of involvementInvolve people formally'̂Wide range of involvemenf^


272621

77

1312

Total

382825108

1413

Did Use

211812

77

78

Mentions by Episode''

Successful

Wish HadUsed

01000

00

Did NotUse

00100

10

Did Use

34

1021

6.3

Unsuccessful

Wisb HadUsed

"114110

01

Did NotUse

31100

01

" Mentions are combined for each interviewee's successful and unsuccessful issues (n = 42 individuals). If an interviewee mentioneda move more than once, it was counted as a single mention.

'' Mentions are separated by successful and unsuccessful episodes (n = 82 episodes) and tallied according to the framing of the mention.If an interviewee did not explicitly or implicitly mention a move as something he or she either used, did not use, or wished he or she hadused, then it was not mentioned at all.

'̂ Not proposed hy Dutton and Ashford (1993).

we discuss later in the section on timing character-istics of the issue-selling process.

Targets of involvement. Knowledge of effectiveissue selling included a sense of whom to involve,not just when involvement should take place. Issuesellers talked about customizing issue-selling ef-forts to include the full range of issue stakeholdersor any solution that was to be attached to the issue.As expected, the most frequently mentioned movewas "involving others at an upper level." Wheninterviewees talked about such others, they werereferring to any person at a higher organizationallevel whom they involved in the issue-selling pro-cess. For example, one interviewee said; "What I'vedone is involved the entire upper-level manage-ment team in defending the position, and so I'vebrought the troops forth." Another commented,describing a failure episode; "I think that if we hadhad more confronting meetings early on with high-er-level management, that would have been bene-ficial." These examples highlight the importancethat sellers placed on involving others at a higherlevel. This tactic was mentioned by more than 60percent of the respondents. Since issue selling, bydefinition, requires making one's case for the sig-nificance of an issue to those higher in an organi-zation's hierarchy than one is oneself, this findingconfirms these managers' belief in tbe utility of themove.

More interesting is that close to half of the issuesellers (48%) mentioned involving those at their

own level or others from somewhere else in theorganization as part of what accounted for a sellingsuccess or failure. Interviewees talked about in-volving departments and individuals. Not surpris-ingly, involving others took effort. As one inter-viewee told us;

Probably the in-depth working with three other de-partments—computer services, personnel, and our-selves. . . . There was an awful lot of coordinationthere that took place. . . . And the only reason wewere successful is because of the cooperation wewere getting between the three departments.

Sellers talked much more ofren about involvingothers at their own levels when they were describ-ing what they did in successful as opposed to un-successful selling episodes. In addition, severalpeople wished they had used this tactic in thefailed selling attempts. Thus, the theory-in-use(Argyris & Schon, 1982) for effective selling seemedto include this involvement tactic as an acceptednecessity.

Half the sellers described involving people fromoutside of the boundaries of the organization. How-ever, there were nearly as many mentions of involv-ing outsiders in successful selling episodes as therewere in unsuccessful selling episodes. In one illus-tration, an issue seller involved an outside personin order to lend credibility; "We also, as a preludeto all of this, engaged an outside consultant, a third-party person, to just give us an opinion about the


MRI field—to try to take it out of the context thatthis is what the chairman of the department wantedor the clinician wanted."

In some issue-selling episodes, sellers mentionedtheir efforts to involve individuals who were lowerthan they were in the hierarchy. Sellers describedusing this tactic three times as ofren in successfulselling stories than in unsuccessful ones. In mostcases, sellers were referring to people who wouldbe affected by action taken if the issue were suc-cessfully sold. One seller explained his failed is-sue-selling attempt; "I guess I thought probablywbere we messed up is not selling it to the employ-ees first, assuming that this was something thatthey were going to like. And we did involve someof them in discussions, but not enough. That'sprobably where a major mistake was made."

Finally, most of the accounts that referred specif-ically to involving the issue seller's boss were aboutsuccessful episodes. Several times, involvement ofthe boss was described as an intermediate step inbuilding the case for an issue. This move resembleswhat Kanter (1983) called "clearing the invest-ment" with your immediate boss, as a type of pre-lude to building a coalition. By advocating tbe is-sue to a supervisor, sellers pursued an incremental,cascading approach to selling an issue upward.

Nature of involvement. Sellers differentiated be-tween formal and informal types of involvement.When they talked about formal involvement, theydiscussed it as the legitimate and sanctioned meansfor selling an issue at Northeast Hospital. For ex-ample, one seller told us about the committees andtask forces that were used as legitimate forums forcreating involvement and interest in issues. He ex-plained, "We did it through committees. We have acouple of committees that deal with operationalissues within our nursing department and else-where." Using formal means of involvement wasmentioned with equal frequency in describing suc-cess and failure.

Also, the talk of involvement included distinc-tions about the range or breadtb of individuals whowere involved in a selling episode. Unlike ourother subcategories, which concern who was in-volved in an issue-selling episode, this subcategoryreflects sellers' mentions of the overall scope of theinvolvement of others in the issue-selling process.For example, one issue seller described her in-volvement tactics as including the creation of a taskforce with diverse membership; "The chairman ofthe task force was the vice president for maternaland child health nursing, and we had representa-tives from finance, from the medical staff, and fromnursing—the intensive care nursery." Sellers men-tioned using broad involvement more ofren in de-

scribing successful as opposed to unsuccessful ep-isodes.

In sum, Dutton and Ashford (1993) emphasizedthe payoffs and costs of involving others, but ourdata focused on whom to involve and how to in-volve them. Involvement moves indicate sensibili-ties about how to create allies to foster issue-sellingsuccess. Effective involvement tactics depend onlaiowing the context (for example, knowing if it isnormative to involve others formally in a givencontext) and having the right social connectionsand knowledge of whom to involve and when toinvolve them.

Process Moves

In this category, Dutton and Ashford (1993) an-ticipated only moves that reflected the formality ofthe process. Interviewees' accounts revealed twoadditional new groupings; preparation and timing.Table 3 gives the frequencies of our process cate-gories and subcategories.

Formality. Issue sellers were thoughtful and de-liberate in their choices about how to proceed withissue selling. One-quarter of the interviewees men-tioned making choices about tbe level of formalityof tbe process. Sometimes the formality of the pro-cess was credited as a positive force in an effectiveselling attempt. Not being formal enough wassometimes viewed as contributing to a process fail-ure. For example, this seller described the reasonsfor a successful attempt; "We followed the proto-cols for packaging an issue. In this case, the pack-aging was in the form of the program summaryprotocol. So we followed the outline and filled inthe text. So we answered the questions that [were]required of us to be answered." Likewise, writtencommunication was a formal process mechanism.The pattern is unclear in its implications for suc-cessful selling.

Preparation. "Doing one's homework" was thecolloquial term for a range of activities throughwhich sellers learned about an issue, any attachedsolutions, and the context of the issue before andduring the issue-selling process. These forms ofpreparation fit nicely with Kanter's account of thebuying-in process used by change advocates. Thisaction served as a type of "check that was designedto screen out chancy or unneeded ideas" (Kanter,1983; 157). For some sellers, preparation involvedcollecting information through meetings with peo-ple inside and outside the organization. For exam-ple, one issue seller described his extensive infor-mation gathering;

I went and met with every single department direc-tor and vice president, etc., individually when I


TABLE 3Process: Formality, Preparation, and Timing


FormalityUse of formal processUse of written communication'^

Preparation'^

TimingPersistence in selling activitiesOpportune timingEarly involvement


1010

17

121612

Total

1314

19

131814

Did Use

58

10

766

Mentions by Episode**

Successful

Wish HadUsed

00

0

0

12

Did NotUse

20

1

002

Did Use

05

8

311

Unsuccessful

Wisb HadUsed

11

0

210

3

Did NotUse

50

0

100

" Mentions are combined for each interviewee's successful and unsuccessful issues [n = 42 individuals). If an interviewee mentioneda move more than once, it was counted as a single mention.

'' Mentions are separated by successful and unsuccessful episodes [n = 82 episodes) and tallied according to the framing of the mention.If an interviewee did not explicitly or implicitly mention a move as something he or she either used, did not use, or wished he or she hadused, then it was not mentioned at all.

" Not proposed by Dutton and Ashford (1993).

arrived here for the first three months. And built abackground of information on the issues that weretroubling the key administrators within the organi-zation, and communications was one of them thatcontinued to come up. . . . So the background wasthere.

Sometimes sellers felt that preparation had en-abled tbem to capitalize on momentum that couldbuild from unexpected places or unanticipated cir-cumstances. One seller described this advantage asfollows; "It takes a good deal of time to get thingsthrough the system, so that by my doing a lot ofpreparation and in setting things up, when I needthings to happen, I've pretty much got people buy-ing into the issue that I'm trying to sell. . . . I can goand get people to act quickly on it." However, thismove was mentioned almost equally in successfuland unsuccessful episodes, suggesting that prepa-ration is relevant but may not be a sufficient con-dition for success.

Timing. Timing was represented in the form ofpersistence, opportunism, and involving others atthe right time. Given that issue-selling processescan take a long time, sellers thought it was impor-tant to be persistent. Most often, respondents char-acterized persistence as a necessary and positivefactor contributing to success in issue selling. Itwas a move mentioned twice as ofren in successfulas in unsuccessful episodes. Frohman (1997) de-scribes change initiators as using persistence tosignal a bias for action. One issue seller describedthe role of persistence thus;

Well, we never let up. And every time we had some-one who was on the floor, we showed the problem,and we also engaged the staff so that staff was con-tinually promoting the idea within the institu-tion. . . . We kept pushing the timeliness of it, and asother institutions undertook this endeavor, we werepromoting the idea internally by sending peoplepublications, etc., surrounding the issue.

Also, sellers' narratives exhibited an astute senseof when it was appropriate and effective to moveforward with an issue and when it might be betterto hold back. Although a number of the successfulaccounts used opportune timing, many of the un-successful accounts indicated that the sellerswished they had used this move. Ofren the sense oftiming capitalized on the salience of an issue. Asone seller indicated; "This is the national trend inmanaging nursing staff. To try to be as flexible asyou can to give the . . . biggest benefit package youcan as a recruitment tool. We've got a nationalnursing sbortage, and so it . . . went along with thetrend."

The selling accounts were replete with a sensethat the timing of selling an issue had some con-trollable ahd some uncontrollable elements. In suc-cessful selling accounts, respondents ofren con-veyed a sense that they knew when to sell an issueon the basis of their assessment of the level ofsupport that existed for the issue, the pressure foraction on it, and the availability of convincingfacts. For sellers, it was ofren a delicate balancing


act of having the forces lined up in the right way toenable the execution of a successful sell.

The final timing question was when to get othersinvolved in the process of issue selling. Sellers'accounts implied that early involvement contrib-uted to the efficiency of the issue-selling attempts.As one interviewee explained; "Involving themearly allowed us to get past that hurdle, and alsosaved me the time of doing it [the issue selling],which was an effective use of staff resources." Sell-ers generally believe that the early involvement ofothers allowed a richer set of cues or signals aboutpossible resistance. Six accounts of successful is-sue selling contain mentions of early involvement.In a handful of episodes, however, the sellers indi-cated retrospectively that they should have usedthis tactic. Thus, although timing of involvement isnot a common move, sellers that mentioned itseemed to indicate that they thought it was useful.

Contextual Knovtrledge and Issue Selling

The data analyses reveal that three kinds of con-textual knowledge are relevant to making issue-selling moves happen. In Table 4 we present thesekinds of knowledge in the form of a set of ques-tions, the answers to which would give the issueseller needed information.

First, issue-selling moves seemed to rely on rela-tional knowledge, or sellers' understanding of theindividuals and the social relationships that wereimportant to their issues. Relational knowledge is

knowledge that "actors have of each other, of eachother's intentions, stakes, private goals and territo-ries" (Baumard, 1999; 22). Knowledge about thesematters helped a seller anticipate and address re-sistance, find and enlist expertise, and learn andwork the power structure. Thus, relational knowl-edge helped sellers navigate the political aspects ofa context. It also helped them to talk about an issuein a way that captured a person's interest. We seethe value of relational knowledge in this seller'saccount of how she customized the sale of an issueto different groups;

It depended on who I was talking to. If I was talkingto [the chief operating officer], I talked about whywe should do it physically. If I was talking to thephysicians, I talked about how it would improvetheir ability to care for their patients. If I was talkingto the nursing staff, I talked about how it wouldimprove their work. I didn't talk about financesbecause I knew they didn't want to hear it.

Several moves implied the importance of rela-tional knowledge. For example, all of the involve-ment moves required relational knowledge, both asense of whom to involve and a sense of how tomotivate their involvement. Sellers who hope to tietheir issues to the concerns of key constituentsmust have relational knowledge to know whichconstituents are key and the nature of their targets'concerns. Even such seemingly unrelated moves asopportune timing and use of written communica-tion can require relational knowledge. For exam-

TABLE 4Contextual Knowledge Important to Issue Selling

Type of Knowledge Questions

Relational

Normative

Strategic

Who will be affected by the issue?Who has experience with the issue?Who cares about the issue?What groups can help with advocating for the issue?What groups might object to this issue?Does this issue threaten anyone or any group?Who has decision authority relevant to the issue?Who has power to promote or to hinder this issue?When will people be ready to hear about this issue?

What kinds of data do people use? In particular, what kinds of data do important people use?How are data normally presented?How are arguments made against an issue?What kinds of protocols are followed?What kinds of meetings or social gatherings are considered legitimate decision forums?How much time does it usually take to sell an issue?Have similar issues been sold (or failed) before?

What are the organization's goals?How does the organization plan to achieve these goals?What are the critical strategic issues for top management?What is our broader competitive context?


pie, part of what makes the timing opportune isknowledge about when the time is right for thetargeted individuals. Similarly, the decision to usewritten rather than spoken communication relieson the relational knowledge of the target personand how he or she likes to receive communication.Smart sellers will use their relational knowledge toguide their moves.

Issue-selling moves also seemed to rely on nor-mative knowledge, or sellers' understanding of theaccepted or appropriate behavior patterns in a par-ticular organizational setting. As Floyd and Wool-dridge stated, "Informal norms are highly idiosyn-cratic to an organization, but knowing them may becritical to smoothing the way for strategic initia-tives" (2000; 201). Ocasio (1997) similarly en-dorsed the importance of normative knowledge. InNortheast Hospital, the normative knowledge con-sidered as relevant to issue selling involved aware-ness of data presentation, protocols, forums, andpast history. We get a sense of how normativeknowledge helps an issue seller in this account;

I really have proposed very little that was not ac-cepted, but it's always inter-institutional, interde-partmental, and I get other department directors tobuy in before I propose it or I don't propose it. . . . Ibelieve in doing your homework before. . . . It helpswhen you know the background of the institutionwell. I think it helps to make decisions of what topursue and what not. Because, I mean, there aremany things that have been tried before, and I mighteven remember why they died, so that helps me inproposing it again.

Another example of the relevance of normativeknowledge concerns the use of formal or informalchannels, which relies on knowledge of organiza-tional norms (Dutton & Ashford, 1993). Respon-dents described their presentations to operatingcommittees and boards of trustees, the formation ofad hoc project teams, and their compliance to pro-cedural hierarchies as all fitting into the standardsof acceptable process at Northeast Hospital. Notusing moves that fell within these standards wasviewed as potentially jeopardizing one's case. Tomake these moves appropriately, however, requiresknowledge of the normative context. Indeed, decid-ing just which moves to make requires normativelaiowledge. Decisions about when to persist, whenis an advantageous time, and when to involve se-lected others all depend on the norms of the situa-tion. The more sellers know about those norms, thebetter able they are to choose moves and executethem well for that specific setting.

Finally, our issue sellers' accounts implied thattheir moves relied on strategic knowledge, or an

understanding of the organization's goals, plans,and priorities. Floyd and Wooldridge (2000) arguedthat loiowledge in this domain, which includes asense of the competitive scene or broader institu-tional context in which an organization is embed-ded, is an important antecedent to strategic re-newal. With such Icnowledge at Northeast Hospital,for example, an issue seller finally sold an issuethat had been evident but ignored for many yearsby invoking external pressures for improved healthcare in order to mobilize people to get behind theissue. As she describes it;

I think I was benefited because I came in at the righttime. If I had tried to sell the concept two years ago,it would not have flown. There's no doubt in mymind. But everybody is ripe for change now. . . .Because they see the changes in health care, andthey realize we can't keep doing the status quo—forjoh security or whatever reason—so they're reallymuch more open for ideas.

Sellers need strategic laiowledge in order to en-gage in most of the bundling moves (for example,tying issues to valued goals) and to present theissues effectively in the logic of the business plan.Although this can be done generically with littlestrategic knowledge, such knowledge will allowsellers to tailor their presentation to the logic oftheir specific organization and its business plan.Taking advantage of opportune timing can also de-pend on strategic knowledge, since what makes atime advantageous so ofren depends on the strate-gic choices and pressures faced by an organizationat that time.

Our data suggest that Westley's (1990) admoni-tion to keep managers actively involved in conver-sations about the strategies of their firms may beparticularly important in organizations that viewissue selling as critical to the firms' adaptive pro-cess (Floyd & Wooldridge, 1996). Being aware ofthe strategic context may facilitate managers' ac-cess to the themes and language necessary for ef-fectively bundling tbe issues with the current goalsof an organization. This knowledge allows sellersto enhance the perceived importance of an issue byconnecting it to goals already deemed important bytop management. Strategic knowledge can alsohelp managers time their moves opportunisticallyto optimize success. Sellers can use pressure pointsand windows of attention to build a case for anissue and to take advantage of the right moments(Scully & Creed, 1998).

DISCUSSION

This study provides a picture of the behind-the-scenes moves that in part compose change pro-


cesses in organizations. It is consistent with effortsto understand how agents "animate change pro-cesses [while] remaining in the shadows" (Scully &Creed, 1998; 24). By analyzing patterns of movesmade in successful and unsuccessful issue selling,we deepen understanding of issue-selling pro-cesses in particular and organizational change pro-cesses more generally.

A Revised View of Issue Selling

The picture of issue selling that emerges fromthese descriptive accounts of success and failuresuggests some support for the moves identified byDutton and Ashford (1993) and uncovers addi-tional categories of choices that sellers make im-plicitly or explicitly when promoting an issue. Ingeneral, the process looks much more political andcontextually embedded than it was portrayed asbeing in the original account. Three data patternssuggest revisions to their model of how issue sell-ing might work in organizations.

First, Dutton and Ashford (1993) suggested thatsellers would be more successful to the extent thatthey sold their issues as strategically important,that the managements of their organizations werecapable of responding, and tbat tbere were payoffsfor the organizations. Our sellers clearly carriedthese three dimensions in their implicit theoriestoo. The sellers' desire to convey strategic impor-tance is evidenced through their tendency to tieissues to important goals. Sellers' concern for man-agement's response capability was evidenced bytheir tendency to break up the issues into bits andpresent tbem incrementally. Concern for showingpayoffs to organizations was indicated through thetendency to use the logic of the business plan inmaking claims about issues. These findings supportOcasio's (1997) contention that decision makersattend to issues with greater organizational legiti-macy, value, and relevance. The issue sellers in thisstudy appeared to pitch their issues in line withthese allocation tendencies.

Second, although there was evidence, as Duttonand Ashford (1993) proposed, that sellers thoughtabout the importance of presentation and bundlingmoves, packaging moves were not as simple astheir original speculations. Rather, a seller's sensethat the issue process must conform to the logic ofa business plan involved a more complex sellingrecipe than simply providing compelling facts. Fol-lowing the logic of a business plan also impliedknowledge of how to present these facts (for exam-ple, in what order and in what form). It was anissue-selling move that advantaged sellers who hadboth normative and strategic knowledge about the

organizational context. However, these patterns ofpackaging moves—using a business logic, continu-ously pushing an issue, packaging an issue as in-cremental as opposed to radical, and tying an issueto valued goals—also imply a conservative bias inthe issue-selling processes in this organization.Sellers may be less likely to raise an issue for whicha business logic is not yet established or that cannotbe broken into small commitments. As a result, onemight expect that without explicit intervention bytop managers, few radical changes would bubbleup from managers below. Our guess is that thisconservative bias, created in part by the conformityto a constrained presentation recipe, is typical ofmany bureaucratic organizational settings.

Third, Dutton and Ashford (1993) also men-tioned moves that did not show up in our data. Forexample, they stated that sellers would be moresuccessful to the extent that they used emotionaland novel terms, told their stories succinctly, usedtwo-sided arguments, and framed them so that topmanagement felt a heightened sense of responsibil-ity for actions. Our interviewees rarely mentionedthese framing moves. It may be that the most salientfeatures of their implicit theories of issue sellingare those that can show the payoff of an issue, andmore subtle moves are less critical. It may also bethat in this culture, the more forceful move of im-plying top management's responsibility was seenas inappropriate.

Three broader patterns are also notewortby. First,sellers' impression management concerns weremuch less salient in these data than implied byDutton and Asbford (1993). That study implied, forexample, that involving others was shaped by im-pression management concerns (for example, if youinvolve others, you'll get less credit). Contrary tothe earlier prediction, sellers at Northeast Hospitalcharacterized involving others mainly as a meansto increase the chances of getting an issue on thetable. Here, as in the discussion of other moves,preservation of a positive self-image was less cen-tral than making moves that fit the political andorganizational context. We suspect that impressionmanagement does play a role in the issue-sellingprocess, but it may be tbat our interviewees thoughtthat admitting to having these concerns to the in-terviewer would be bad impression management!Such concerns have been shown to be important inpast qualitative (Dutton et al., 1997) and quantita-tive (Ashford et al., 1998) research on the initiationof issue selling. Alternatively, it may be that im-pression-management concerns affect choices tosell or not sell an issue but have less of an impacton the selling process itself.

Second, issue selling as revealed by this study is


much more of a political and commitment-buildingprocess than Dutton and Ashford originally con-ceived it to be. Such a view supports Kanter's(1983) original account of change masters in organ-izations and Floyd and Wooldridge's (2000) morerecent description of middle management's role inthe strategic renewal process. The process is alsodeeply contextual in the sense of requiring issuesellers to have in-depth knowledge of "how thesystem works" in order to effectively make changehappen. For example, the selling stories uncoverthe knowledge required to execute the involvementmoves needed to build support for an issue. Al-though Dutton and Ashford (1993) speculated thatthe key issue-selling choices around involvementwere whether one went solo or not, from this study,we see that involvement moves offer choices aboutwhom to involve, at what level, at what time, andin what manner. In addition, the importance ofissue bundling implies astute issue sellers bave agood sense of the kinds of concerns and targets ofattention that already have currency in their organ-ization. This knowledge requires political acumenand a willingness to act on it. The overall picturesuggests that achieving change through issue sellingis closely tied to how effectively sellers orchesfrategetting the right people involved at the right time andin the right way. Such effectiveness is a form of prac-tical intelligence that has utility for these bottom-upchange efforts but also relies on a combination ofrelational, normative, and sfrategic knowledge.

Third, issue-selling processes are in part shapedby a seller's sense of tbe timing of such efforts. Thecategory of timing-related moves was unantici-pated by Dutton and Ashford (1993) as an impor-tant aspect of issue selling, yet these sellers clearlyactivated change efforts through the timing of dis-cretionary moves. Sellers can, with varying degreesof success, read the contextual signs indicatingreadiness to commit to, act upon, or consider anissue. Their efforts to establish the relevance of anissue suggest that temporal sorting processes exem-plified by tbe "garbage can model" may actually beformed by individuals' attention-allocation efforts(March, 1994). Bundling an issue is an example ofthe way that issue sellers instrumentally shapechoice opportunities in an otherwise disorderlysystem. What we observed in this setting was verysimilar to Kingdon's (1990) characterization of ef-fective docket setters in Congress as good "surfers,"people waiting patiently and sensing when the con-ditions are right to "ride the wave" that will get anissue on the agenda.

Individuals' efforts to create change are tempo-rally embedded (Dacin, Ventresca, & Beal, 1999).Our findings are consistent with the idea that there

are event-triggered windows of opportunity formaking change happen (Cersick, 1994; Tyre &Orlikowski, 1994). Astute issue sellers scan theirenvironment for nonroutine events via which theycan effectively mobilize attention for issues. Prep-aration and persistence ready the seller and thetarget before such an opportunity. These findingsare supported by research that shows that the tim-ing of key adaptive behaviors distinguishes high per-formers from low performers (Waller, 1999). Our re-search also suggests that the timing of change effortsmay have as much to do with middle managers as itdoes with top executives, who are typically seen asthe sole instigators of organizational reorientation(Tushman & Romanelli, 1985). Thus, sellers' effortscan be critical at special moments by affecting theonset of change efforts (Scully & Creed, 1998).

Future research would be well served if therewere greater attention to how sellers (or other peo-ple taking discretionary action in organizations)learn to make their moves or sense that the timingfor action (or inaction) is right. These sensibilitieswere articulated in this study, but how such aware-ness is created remains a mystery. For example,studies could investigate how organization mem-bers acquire sensibilities about appropriate timingor about their organization's political environmentand how these forms of practical intelligence affectdiscretionary processes. Consideration of thesequestions may reveal how issue sellers entrain tointernal and external rhythms (Ancona & Chong,1996), such as legal and regulatory deadlines, me-dia events, and other patterned regularities. Also,researchers involved in such future efforts couldconsider how irregularly timed aspects of an inter-nal organizational context (such as other issues,limited resources, and "shocks to the system") andthe external environment affect sellers' thinkingand success in timing their issue-selling moves. Itis also important to determine if these moves arecombined in different ways to form systematic pat-terns of timing moves that coincide with the out-comes of selling behavior.

The Importance of Knowledge for Navigating theOrganization Context

Three types of process laiowledge underlie man-agers' implicit theories about issue selling; rela-tional, normative, and strategic. These types of em-bodied and embedded (Blackler, 1995) knowledgeabout "the way things work around here" help toexplain why certain moves during issue sellingmay have been viewed as more or less successful.Relevant issue-selling knowledge involves knowl-edge about how to proceed, how to involve others.


and so forth (Ryle, 1949) that is tied to knowledgeabout an organization's relational, normative, andstrategic contexts.

Knowledge about the "how" of issue selling re-quires both skill in interpretation and a "feel for thegame" being played inside an organization. It forcessellers to recognize that they have little control overmany larger aspects of the organization but muchcontrol over small but important things. For exam-ple, issue sellers do have control over whom theyinvolve, the timing of their moves, and their ownpersistence. Through these types of moves, theyindirectly control the process by being ready to actwith others when the time is right. One intervieweeconveyed this insight in this way;

As long as you are patient enough, and you have agood idea, and something is important, it will hap-pen. You just have to he patient. If I'd stayed in thatposition, I wouldn't have given up on it. I just wouldhave let it rest for about six months and then wouldhave started stirring it again.

Two practical implications flow from this knowl-edge-based perspective. First, individuals whohave deeper and broader knowledge about a rela-tional, normative, and strategic context will be bet-ter equipped to execute a variety of issue-sellingmoves. Tbey can use it to tailor issue-selling effortsto what best fits a particular organizational setting.Although our particular data collection methodol-ogy did not allow us to view the sellers over time,we would expect that star sellers would emerge—managers with a track record of issue-selling suc-cess. That success, we suspect, would be basedupon their contextual knowledge and the interper-sonal skills necessary to navigate the context effec-tively.

Second, top management can facilitate thesetypes of navigation and orchestration efforts by en-bancing managers' opportunities to acquire and up-date their relational, normative, and strategicknowledge. Socialization programs, attempts to in-volve managers at all levels in strategic conversa-tions (Westley, 1990; Wooldridge & Floyd, 1990),and cultural audits are just some of the ways thatmanagers acquire and sense these critical pieces ofcontextual knowledge. By failing to make sure thatpeople are equipped with the knowledge to achieveeffective selling from below, organizations robthemselves of this vital source of internal changeinitiatives.

Implications for OrganizationalChange Processes

Our study of issue selling at Northeast Hospital isa study of the initial stage of change—that of focus-

ing attention on an issue. It contributes to thechange literature in organizations in several ways.

First, consistent with research showing the dom-inance of incremental over radical organizationalchange (e.g., Tushman & Romanelli, 1985), themoves evidenced in Northeast Hospital suggest aconservative change pattern. The repertoire ofissue-selling moves found here suggests that, incases of issue selling occurring in typical, largebureaucratic organizations and without interven-tion by top management, Burgelman's (1983, 1991)portrayal of most change efforts as fitting withrather than challenging the scope of an organiza-tion's strategy may be appropriate. Even morenovel issues, such as cross-training support ser-vices staff and computerizing the operating room,were characterized as being sold successfully be-cause they were toned down and made to appear tobe more incremental changes through packagingand involvement tactics. This portrait of middle man-agers is consistent with Jackall's (1988) observationthat middle managers can be a conservative groupwho avoid nonconformity in order to get ahead.

Second, issue selling is consistent with Weickand Quinn's (1999) characterization of change ascontinuous rather than episodic. Continuouschange tends to be "ongoing, evolving, and cumu-lative" (Weick & Quinn, 1999; 375). Those authorsargued that with continuous change, translationprocesses are key, where ideas (or, in our case,issues) get matched to purposes and localities(Czarniawska & Jeorges, 1996). Issue selling may bean important means of translation in continuouschange efforts; sellers actively try to present andtime initiatives and to involve others in ways thatfit the context and do not violate existing normsand practices. The imperfections in this translationprocess, and the amount of time and specificity ofcircumstances needed to make it work, mean thatthese efforts go on all the time. As Cohen, March,and Olsen (1972) proposed in their "garbage canmodel," when change happens, it is ofren becauseof a fortuitous convergence of issues, solutions, andopportunities, an event an issue seller can onlyindirectly control but that smart sellers deliberatelytry to control. Future research might profitably con-cern how the issue-selling efforts of multiple man-agers can intersect and collide to promote or im-pede change.

Third, the dominant portrait of change agents hascentered on the heroic efforts of those at the tops oforganizations or those called in to make changefrom the outside (consultants). An issue-sellingview is part of a growing trend (Floyd & Wool-dridge, 2000; Morrison & Phelps, 1999) toward rec-ognizing the less visibly beroic, behind-the-scenes


work and efforts of those below the top manage-ment group. In fact, issue sellers are mindful andproactive orchestrators of change (Kanter, 1983)who, with intent, try to bring streams of issues,solutions, and opportunities together in ways thatfocus attention and invite action on issues. Issuesellers' actions are evidence of the entrepreneurialresources that individuals use to affect the activa-tion and direction of change efforts (Mosakowski,1998). Issue sellers exercise discretion in how theyconstruct an issue (Pitt, MacAuley, Dowds, & Sims,1997), as well as in how they mobilize involvementand attentional investment in it. This portrait callsfor an expansion of the change agent skills re-searchers and others consider relevant and de-scribes a new target for skill-building efforts; themiddle manager. Indeed, making change from be-low requires specific types of practical knowledgeand skill. We agree with Ford and Ford (1995) thatchange takes place in conversations, and our re-search suggests that future work should focus onthe broad set of skills and competencies needed forthis kind of behind-the-scenes work.

Finally, our study suggests a complementary setof moves that top managers might make to promoteissue selling and change in their organizations. Thebest top managers will not only be mindful aboutthe cultures they create, the enabling structurestbey put into place, and the resources they allocate,but will also be mindful about how these factorscontribute to patterns of change that occur overtime. This fundamental shifr leaves to top managersthe important job of constructing the context byshaping the norms and structures that promote thebehavior they want. One example might be theestablishment of what Brown and Eisenhardt(1997) termed "semistructures" that fix some as-pects of context (for example, responsibilities,project priorities) but give great freedom on otheraspects. The freedom allows organizations to capi-talize on the creativity and change potential of mid-dle managers, whereas the partial fixing allows forcoordination. Thus, both organizations and indi-viduals can develop repertoires of skill-based ac-tions that influence what changes can happen aswell as how those changes happen in organizations(Sitkin, Sutcliffe, & Weick, 1998).

Limits of the Study

This kind of study has several limitations. First,our data come from a single organizational contextthat has been bistorically isolated from extensivecompetitive pressures. All of the managers whomwe interviewed described Northeast Hospital as afairly bureaucratic and formal context, and the

issue-selling accounts obviously reflect thesecontextual features. Thus, our claims about howmanagers think about issue-selling moves, and thetheoretical implications derived from this empiri-cal pattern, are most easily generalized to similarorganizational settings. Future research should in-clude a greater variety of organizational contexts tobetter explain how issue-selling knowledge is ac-quired and how it works.

Second, our data consist of retrospective ac-counts. One problem with such accounts is theaccuracy of the recall of events and processes. Wealso have no way to determine whether it was theway sellers actually executed moves that affectedfailure or success. Here, as in previous studies (e.g.,Dutton et al., 1997), managers had no trouble re-membering or describing their issue selling (sug-gesting that they are reflective about this behavior),but we cannot assess the accuracy of their accounts.Another problem may be the reconstruction ofevents with positive (successful) rather than nega-tive (unsuccessful) outcomes, owing to self-servingattribution biases and other factors (Fiske & Taylor,1984). In these data, it is difficult to distinguish talkabout real moves from normative talk about failedand successful moves. We tried to minimize thisdifficulty by being cautious in interpreting findingswhen self-aggrandizement was a plausible alterna-tive explanation. However, this limit remains andwill need to be addressed in future work on issueselling.

Third, our interview guide asked sellers to reflecton issue-selling episodes involving strategic or im-portant issues for their organization. Patterns ofmoves could be very different for issues thought tobe less strategic. Given our focus, our emphasis onstrategic issues makes sense, but it may limit thegeneralizability of our results.

CONCLUSION

From a narrow perspective, this research contrib-utes to understanding tbe processes through whicha manager's initiatives can shape top management'sattention. More broadly, our research begins to un-ravel and make sense of the microprocesses thatcompose strategic change (e.g., Floyd & Wool-dridge, 1996; Ghoshal & Bartlett, 1994; Noda &Bower, 1996; Westley, 1990). In this way, our studyworks at the interface between organizational be-havior and corporate strategy.

In a pluralistic organizational world, managerscoexist witb different and competing interests andperspectives. Through the issue-selling process,managers push their ideas forward to effect change.Our issue-selling accounts reveal the importance of


relational, normative, and strategic contextualknowledge in this process. This situated, pragmaticlaiowledge enables situated, competent action(Blackler, 1995; Lave, 1993; Star, 1992). Successfulsellers drawing on this knowledge employ a varietyof moves to achieve their issue-selling objectives,including characterizing their appeal as a rational,incremental step, tied to important organizationalgoals and priorities. They involve a wide range ofothers, early in the process, with a special focus oninvolving those at their level or above. They keeptheir bosses informed, and they persist in theirselling efforts while taking advantage of timing todecide when to sell and when to hold back. Inshort, managers have nuanced theories about howto act successfully in their contexts. This studysuggests that it is important to understand howmanagers read and navigate their strategic andstructural contexts in order to benefit tbemselvesand their organizations. Indeed, we propose thatsenior management's context design mandate (e.g.,Btu-gelman, 1994; Choshal & Bartlett, 1994; Noda &Bower, 1996) can only be accomplished if it is basedon an understanding of the thought patterns of thosewho are trying to "work" the context. Theirs are themoves that matter; moves that will serve as internalengines for change and innovation. Given their im-portance, these moves are worth cultivating throughexplicit senior management attention and interven-tion to creating a context that effectively mobilizesthis bottom-up change process.

REFERENCES

Ancona, D., & Chong, C. L. 1996. Entrainment: Pace,cycle, and rhythm in organizational behavior. InB. M. Staw & L. L. Cummings (Eds.), Research inorganizational behavior, vol. 18; 251-284. Green-wich, CT; lAI Press.

Ansoff, H. 1.1980. Strategic issue management. StrategicManagement Journal, 1: 131-148.

Argyris, C, & Schon, D. 1982. Theory in practice. SanFrancisco: lossey-Bass.

Ashford, S., Rothbard, N. P., Piderit, S. K., & Dutton, ). E.1998. Out on a limb: The role of context and impres-sion management in selling gender-equity issues.Administrative Science Quarterly, 43: 23-57.

Barry, D., & Elmes, M. 1997. Strategy retold: Toward anarrative view of strategy discourse. Academy ofManagement Review, 22: 428-452.

Baumard, P. 1999. Tacit knowledge in organizations.London: Sage.

Blackler, F. 1995. Knowledge, knowledge work and or-ganizations: An overview and interpretation. Organ-ization Studies, 16: 1021-1046.

Boje, D. 1991. The storytelling organization: A study ofstory performance in an office-supply firm. Admin-istrative Science Quarterly, 36: 106-126.

Bourdieu, P. 1977. Outline of a theory of practice. NewYork: Cambridge University Press.

Brown, S. L., & Eisenhardt, K. M. 1997. The art of con-tinuous change: Linking complexity theory andtime-paced evolution in relentlessly shifting organi-zations. Administrative Science Quarterly, 42:1-34.

Bruner, J. 1990. Acts of meaning. Cambridge, MA:Harvard University Press.

Burgelman, R. A. 1983. A process model of internal cor-porate venturing in the diversified major firm.Administrative Science Quarterly, 28: 223-224.

Burgelman, R. A. 1991. Intraorganizational ecology ofstrategy making and organizational adaptation: The-ory and field research. Organization Science, 2:239-262.

Burgelman, R. A. 1994. Fading memories: A process the-ory of strategic business exit in dynamic environ-ments. Administrative Science Quarterly, 39: 24-56.

Burgelman, R. A., & Sayles, L. R. 1986. Inside corporateinnovations: Strategy and managerial skills. NewYork: Free Press.

Cohen, M. D., March, ). G., & Olsen, ). P. 1972. A garbagecan model of organizational choice. AdministrativeScience Quarterly, 17: 1-25.

Collins, R. 1981. On the microfoundations of macrosoci-ology. American Journal of Sociology, 86: 984-1013.

Creed, W. E. D., & Scully, M. A. 2000. Songs of ourselves:Employees' deployment of social identity in work-place encounters. Journal of Management Inquiry,9: 391-412.

Czarniawska, B. 1997. A narrative approach to organi-zation studies. Thousand Oaks, CA: Sage.

Czarniawska, B., &Ieorges, B. 1996. Travels of ideas. In B.Czarniawska & G. Sevon (Eds.), Translating organi-zational change: 13-48. Berlin: de Gruyter.

Dacin, M. T., Ventresca, M. )., & Beal, B. D. 1999. Theembeddedness of organizations: Dialogue & direc-tions' Journal of Management, 12: 243-259.

Dean, ). W. 1987. Building the future: The justificationprocess for new technology. In ). M. Pennings & A.Buitendam (Eds.), New technology as organiza-tional innovations: 35-38, Cambridge, MA: Ball-inger.

Dutton, ). E. 1997. Strategic agenda building in organiza-tions. In Z. Shapira (Ed.), Organizational decisionmaking: 81-107. Cambridge, England: CambridgeUniversity Press.

Dutton, I. E., & Ashford, S. ). 1993. Selling issues to topmanagement. Academy of Management Review,18: 397-428.


Dutton, ). E., Ashford, S. )., Wierba, E. E., O'Neill, R., &Hayes, E. 1997. Reading the wind: How middle man-agers assess the context for issue selling to top man-agers. Strategic Management Journal, 15: 407-425.

Dutton, J. E., & Webster, ). 1988. Patterns of interestaround issues: The role of uncertainty and feasibil-ity. Academy of Management Journal, 31: 663-675.

Finkelstein, S., & Hamhrick, D. C. 1996. Strategic lead-ership: Top executives and their effects on organi-zations. Minneapolis: West.

Fiske, S. T., & Taylor, S. E. 1984. Social cognition. NewYork: Random House.

Fligstein, N. 1997. Social skill and institutional theory.American Behavioral Scientist, 40: 397-405.

Floyd, S. W., & Wooldridge, B. J. 1996. The strategicmiddle manager. San Francisco: lossey-Bass.

Floyd, S. W., & Wooldridge, B. ]. 2000. Building strategyfrom the middle: Reconceptualizing strategy pro-cess. Thousand Oaks, CA: Sage.

Ford, ). D., & Ford, L. W. 1995. The role of conversationsin producing intentional change in organizations.Academy of Management Review, 20: 541-566.

Frohman, A. L. 1997. Igniting organizational change frombelow: The power of personal initiative. Organiza-tional Dynamics, 25(3): 39-53.

Gersick, C.). G. 1994. Pacing strategic change: The case ofa new venture. Academy of Management Journal,37: 9-45.

Ghoshal, S., & Bartlett, C. A. 1994. Linking organizationalcontext and managerial action: The dimensions ofquality of management. Strategic ManagementJournal, 15: 91-112.

Goffman, E. 1981. Forms of talk. Philadelphia: Univer-sity of Pennsylvania Press.

Golden, B. R. 1992. The past is the past—Or is it? The useof retrospective accounts as indicators of past strat-egy. Academy of Management Journal, 35: 848-860.

Hambrick, D., & Mason, P. 1984. The upper echelons:The organization as a reflection of its top managers.Academy of Management Review, 9: 19^-206.

Hambrick, D. C. 1989. Putting top managers back in thestrategy picture: Introduction. Strategic Manage-ment Journal, 10: 5-15.

Harquail, C. V. 1996. When one speaks for many: Theinfluence of social identification on group advo-cacy in organizations. Unpublished doctoral disser-tation. University of Michigan, Ann Arbor.

Howell, J. M., & Higgins, C. A. 1990. Champions of tech-nological innovation. Administrative Science Quar-terly, 35: 317-341.

lackall, R. 1988. Moral mazes: The world of corporatemanagers. New York: Oxford University Press.

Kanter, R. M. 1983. The change masters. New York:Simon & Schuster.

Kingdon,). W. 1990. How do issues get on public policyagendas? Paper presented at the annual meetingof the American Sociological Association, Washing-ton, DC.

Kipnis, D., & Schmidt, S. M. 1983. An influence perspec-tive on bargaining within organizations. In M. H.Bazerman & R. ). Lewicki (Eds.), Negotiating in or-ganizations: 303-319. Beverly Hills, CA: Sage.

Klein, G. 1998. Sources of power: How people makedecisions. Cambridge, MA: MIT Press.

Lafigley, A., Mintzberg, H., Pitcher, P., Posada, E., &Saint-Macary, ). 1995. Opening up decision making:The view from the black stool. Organization Sci-ence, 6: 260-279.

Lave, ). 1988. Cognition in practice: Mind, mathematicsand culture in everyday life. Gambridge, England:Cambridge University Press.

Lave, J. 1993. The practice of learning. In S. Ghaikin & ).Lave (Eds.), Understanding practice: Perspectiveson activity and practice: 3-35. Cambridge, England:Cambridge University Press.

Lewin, K. 1951. Field theory in social science. NewYork: Harper & Row.

Lyles, M. A., & Reger, R. K. 1993. Managing for autonomyin joint ventures: A longitudinal study of upwardinfluence. Journal of Management Studies, 30:383-404.

March, ). G. 1994. A primer on decision making: Howdecisions happen. New York: Free Press.

March, I. G., & Olsen, I. P. 1976. Ambiguity and choice inorganizations. Bergen, Norway: Universitetsfor-laget.

March, ). G., & Shapira, Z. 1982. Behavioral decisiontheory and organizational decision theory. In G. R.Ungson & D. N. Braunstein (Eds.), Decision making:An interdisciplinary inquiry: 95-122. Boston: Kent.

Martin, )., Feldman, M. S., Hatch, M. )., & Sitkin, S. B.1983. The uniqueness paradox in organizational sto-ries. Administrative Science Quarterly, 28: 438-453.

Meyerson, D., & Scully, M. 1995. Tempered radicals andthe politics of ambivalence and change. Organiza-tional Science, 6: 585-600.

Miceli, M. P., & Near, ). P. 1992. Blowing the whistle.New York: Lexington Books.

Miles, M. B., & Huberman, A. M. 1994. Qualitative dataanalysis (2nd ed.). Thousand Oaks, CA: Sage.

Mintzberg, H. 1997. The manager's job: Folklore and fact.In R. P. Vecchio (Ed.), Leadership: Understandingthe dynamics of power and influence in organiza-tions: 35-53. Notre Dame, IN: University of NotreDame Press.


Morrison, E. W., & Phelps, C. C. 1999. Taking charge atwork: Extrarole efforts to initiate workplace change.Academy of Management Journal, 42: 403-419.

Mosakowski, E. 1998. Entrepreneurial resources, organi-zational choices, and competitive outcomes. Organ-ization Science, 9: 625-643.

Murnighan, J. K. 1978. Models of coalitional behavior:Game-theoretic, social psychological and politicalperspectives. Psychological Bulletin, 85: 1130-1153.

Noda, T., & Bower, J. L. 1996. Strategy making as iteratedprocesses of resource allocation. Strategic Manage-ment Journal, 17: 159-192:

Nonaka, I. 1994. A dynamic theory of organizationalknowledge creation. Organization Science, 5: 14-37.

Ocasio, W. 1997. Toward an attention-based view of thefirm. Strategic Management Journal, 18: 187-206.

O'Connor, E. S. 1997. Telling decisions: The role of nar-rative in organizational decision making. In Z. Sha-pira (Ed.), Organizational decision making: 304-323. New York: Gambridge University Press.

O'Connor, E. S. 1999. Pluralism in the field: Narrativestudies as an interdisciplinary inquiry. Paper pre-sented at the annual meeting of the Academy ofManagement, Chicago.

Pentland, B. 1992. Organizing moves in software supporthot lines. Administrative Science Quarterly, 37:527-548.

Pfeffer, ). 1992. Managing with power: Politics and in-fluence in organizations. Boston: Harvard BusinessSchool Press.

Pitt, M., MacAulay, L., Dowds, N., & Sims, D. 1997. Horseraces, governance and the chance to fight: On theformation of organizational agendas. British Journalof Management, 8: S19-S30.

Ryle, G. 1949. The concept of mind. London: Hutchin-son's University Library.

Schriesheim, C. A., & Hinkin, T. R. 1990. Influence tac-tics used by subordinates: A theoretical and empiri-cal analysis and refinement of the Kipnis, Schmidtand Wilkinson suhscale. Journal of Applied Psy-chology, 75: 246-257.

Schwenk, C. R. 1985. The use of participant recollectionin the modeling of organizational decision pro-cesses. Academy of Management Review, 10: 496-503.

Scully, M. A., & Creed, W. E. D. 1998. Switch persons onthe tracks of history: Situated agency and con-tested legitimacy in the diffusion of domestic part-ner benefits. Paper presented at the annual meeting sof the Academy of Management, San Diego.

Shortell, S. M., Morrison, E. M., & Friedman, B. 1990.Strategic choices for America's hospitals. SanFrancisco: lossey-Bass.

Simon, H. 1947. Administrative behavior: A study of

decision-making processes in administrative organ-izations. New York: Macmillan.

Sitkin, S. B., Sutcliffe, K. M., & Weick, K. E. 1998. Organ-izational learning. In R. Dorf (Ed.), The technologymanagement handbook: 770-776. Boca Raton, FL:CRC Press.

Star, S. L. 1992. The Trojan door: Organizations as workand the open "black box." Systems Practice, 5: 395-410.

Stumpf, S. A., & Mullen, T. P. 1992. Taking charge:Strategic leadership in the middle game. Engle-wood Cliffs, NI: Prentice-Hall.

Tushman, M., & Romanelli, E. 1985. Organizational evo-lution: A metamorphosis model of convergence andreorientation. In L. L. Cummings & B. M. Staw (Eds.),Research in organizational behavior, vol. 7: 171-222. Greenwich, GT: lAI Press.

Tyre, M. ]., & Orlikowski, W. 1994. Windows of oppor-tunity: Temporal patterns of technological adapta-tion in organizations. Organization Science, 5: 98 -118.

Wagner, ). A. 1994. Participation's effects on perfor-mance and satisfaction: A reconsideration of re-search evidence. Academy of Management Re-view, 19: 312-330.

Wagner, R. K., & Sternberg, R. ). 1985. Practical intelli-gence in real-world pursuits: The role of tacit knowl-edge. Journal of Personality and Social Psychol-ogy, 49: 436-458.

Waller, M. ). 1999. The timing of adaptive group re-sponses to nonroutine events. Academy of Manage-ment Journal, 42: 127-137.

Weick, K. E. 1984. Small wins: Redefining the scale ofsocial problems. American Psychologist, 39: 40-49.

Weick, K. E. 1995. Sensemaking in organizations.Thousand Oaks, CA: Sage.

Weick, K. E., & Browning, L. 1986. Argument and narra-tion in organizational communication. Journal ofManagement, 12: 243-259.

Weick, K. E., & Quinn, R. E. 1999. Organizational changeand development. In ]. T. Spence, ]. M. Darley, & D.).Foss (Eds.), Annual review of psychology, vol. 50:361-386. Palo Alto, CA: Annual Reviews.

Weiss, I. 1989. The powers of problem definition: Thecase of government paperwork. Policy Sciences, 22:97-121.

Westley, F. 1990. Middle managers and strategy: Micro-dynamics of inclusion. Strategic ManagementJournal, 11: 337-352.

Wooldridge, B., & Floyd, S. W. 1990. The strategy pro-cess, middle management involvement, and organi-zational performance. Strategic Management Jour-nal, 11: 231-241.


APPENDIX

Category Definitions Used in Issue-Selling Episodes

1. PackagingUse logic of business plan: The degree to which the

issue seller packaged the issue in a way that is similarto a business plan. The issue seller might have con-centrated on financial aspects ("running the num-bers"), marketing, strategy, operations, or a combina-tion of business perspectives.

Make continuous proposals: The issue seller eithermade a single proposal/plan or presented continuousproposals in succession.

Package issue as incremental: The issue was pack-aged by the issue seller using a step-by-step process tosell the issue. This incremental packaging involvedpresenting different aspects of the issue one piece at atime until the complete story was told.

Tie issue to valued goals—Profitability: The issuewas linked to monetary consequences or profitability.

Tie issue to valued goals—Market share/organiza-tional image: The issue was linked to market forces orto the image the organization wanted to project.

Tie issue to concerns of key constituents: The issuewas linked to the customers of the organization—thepatients.

Tie issue to other issues: The issue was linked to otherissues that already had been sold as well as to otherissues that were currently in various stages of being sold.

2. InvolvementInvolve others at higher level: The level of the in-

volvement (from weak to strong) of others at levels abovethe issue seller.

Involve others at same level: The level of involvement(from weak to strong) of others at the same level as theissue seller.

Involve others outside organization: The level of in-volvement of parties outside the hospital, including or-ganizations or people who would affect the success ofthe issue-selling process and those who would be im-pacted by the issue-selling process if it were successful.These parties included the community, the state, insur-ance companies, and the health care industry as a whole.

Involve others at lower level: The level of involve-ment (from weak to strong) of others at levels below theissue seller.

Keep boss informed: The issue seller determined theextent to which he or she involved his or her boss in theissue-selling process by means of keeping the boss in-formed of what was going on during the selling process.

Involve people formally: This defined whether theissue-selling process and/or issue seller were endorsedor sanctioned by the institution by means of a task forceor official endorsement.

Wide range of involvement: The issue seller saw in-volvement as broad, with the administration as well asphysicians involved.

3. ProcessUse of formal process: The degree to which formal

organizational processes were followed and observedduring the selling process.

Use of written communication: The amounts ofmemos, letters, proposals, and other written forms ofcommunication that were used in the selling process.

Preparation: The amount of research and investigat-ing (or homework) that an issue seller did before andduring the selling process.

Persistence in selling activities: The extent to whichthe issue seller used constant pressure to sell the issue.

Opportunistic timing: The availability of an opportu-nity to start the selling process.

Early involvement (timing of involvement): The issueseller identified the point (for example, early in the pro-cess) at which she or he or others became involved in theissue-selling process. Typically, early involvement re-ferred to being involved at the beginning of the issue-selling process.

Jane E. Dutton is the William Russell Kelly Professor ofBusiness Administration at the University of MichiganBusiness School. She received her Ph.D. from Northwest-ern University. Her interests focus on relational accountsof individuals' work experiences. She is studying feel-ings and interactions, compassion, and invisible rela-tional work in organizations.

Susan J. Ashford received her Ph.D. from NorthwesternUniversity. She is the Michael and Susan Jandernoa Pro-fessor of Business Administration at the University ofMichigan. Her interests focus on the ways in which in-dividuals are proactive in organizational life, particularlyon how proactive behavior is an important input to stra-tegic processes.

Regina M. O'Neill is an assistant professor of manage-ment at Suffolk University. She received her Ph.D. fromthe University of Michigan Business School. Her inter-ests focus on professional and personal relationships atwork and the role these relationships play in mentoring,social support, leadership, career development, and up-ward influence processes.

Katherine A. Lawrence is a doctoral student in the De-partment of Organizational Behavior at the University ofMichigan Business School. She received an Ed.M. fromHarvard University and a B.A. from Yale University. Herresearch interests include improvisation, creativity, andhow individuals act with skill.

moves that matter: issue selling and...

Documents