CONSTRUCTING MARKETS AND SHAPING BOUNDARIES: ENTREPRENEURIAL POWER IN NASCENT FIELDS FILIPE M. SANTOS INSEAD KATHLEEN M. EISENHARDT Stanford University We examine how entrepreneurs shape organizational boundaries and construct mar- kets through an inductive, longitudinal study of five ventures. Our central contribution is a framework of how successful entrepreneurs attempt to dominate nascent markets by co-constructing organizational boundaries and market niches using three processes: claiming, demarcating, and controlling a market. We propose that power is the un- derlying boundary logic and indicate the “soft-power” strategies by which entrepre- neurs compete in highly ambiguous markets. Overall, we develop a holistic view of organizational boundaries and offer insights into institutional entrepreneurship and resource dependence theories. Our most important contribution is reinvigorating the study of interorganizational power. Organizational boundaries are fundamental. Ev- ery organization needs to establish its boundary to distinguish the organization from the environment and define its domain of action (Aldrich & Ruef, 2006; Scott, 2003). Given this central role, the phe- nomenon of organizational boundaries has been addressed with a set of rich theoretical perspectives (Santos & Eisenhardt, 2005). A first line of research adopts an exchange efficiency view, looking at cost minimization as a key driver of boundaries (Dyer, 1996; Williamson, 1981, 1991). In a second stream, organizational boundaries are examined through a power lens, with a focus on how organizations can control their exchange relations (Pfeffer & Salancik, 1978; Thompson, 1967). Other research relies on a competence view, in which the evolving resources and capabilities of organizations shape their boundaries (Brusoni, Prencipe, & Pavitt, 2001; Pen- rose, 1959; Peteraf, 1993). A fourth line of research adopts an identity perspective, focusing on the cog- nitive frames of organization members that define “who we are” as an organization and shape the choice of boundaries (Dutton & Dukerich, 1991; Porac, Thomas, Wilson, Paton, & Kanfer, 1995; Tripsas, 2009). Yet, in contrast to the richness of these theoreti- cal perspectives, the range of research designs for studying organizational boundaries has been nar- row. Indeed, much empirical research represents an atomistic view that focuses primarily on the antecedents of single-boundary decisions in cross- sectional samples (David & Han, 2004). The origin of the atomistic view lies primarily in the early influence of exchange efficiency conceptions of boundaries, notably transaction cost economics (TCE), for which the canonical problem is whether to internalize or outsource a specific transaction (Williamson, 1985). This problem formulation has led to the dominant approach of analyzing organi- zational boundaries as discrete structural alterna- tives such as make or buy. As a result, much em- pirical research examines boundary decisions in isolation, as if the shaping of organizational bound- aries were simply the accumulation of independent boundary decisions in well-structured settings. Al- though research has moved from efficiency-based theoretical explanations to include other decision drivers, such as competencies (Argyres, 1996; Jaco- bides & Hitt, 2005; Poppo & Zenger, 1998), the earlier mind-set of focusing on independent bound- We gratefully acknowledge the research support pro- vided by the Stanford Technology Ventures Program and INSEAD R&D. We appreciate the thoughtful comments of Charles Galunic, Morten Hansen, Sarah Kaplan, Riitta Katila, Woody Powell, Ray Levitt, Rory McDonald, Huggy Rao, Kaye Schoonhoven, Wes Sine, Bob Sutton, Patricia Thornton, and seminar participants at Stanford University, INSEAD, the University of Colorado, the Uni- versity of California, Berkeley, Harvard University, the University of Maryland, and Oxford University; and at the Queen’s University Alliance Edge Conference, Prince Bertil Symposium, West Coast Research Symposium on Technology Entrepreneurship, Atlanta Competitive Ad- vantage Conference, and inaugural LBS Entrepreneur- ship Conference. We also acknowledge the valuable con- tributions of Sara Rynes-Weller and three anonymous AMJ reviewers. Academy of Management Journal 2009, Vol. 52, No. 4, 643–671. 643 Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s express written permission. Users may print, download or email articles for individual use only.
Transcript
CONSTRUCTING MARKETS AND SHAPING BOUNDARIES:ENTREPRENEURIAL POWER IN NASCENT FIELDS
FILIPE M. SANTOSINSEAD
KATHLEEN M. EISENHARDTStanford University
We examine how entrepreneurs shape organizational boundaries and construct mar-kets through an inductive, longitudinal study of five ventures. Our central contributionis a framework of how successful entrepreneurs attempt to dominate nascent marketsby co-constructing organizational boundaries and market niches using three processes:claiming, demarcating, and controlling a market. We propose that power is the un-derlying boundary logic and indicate the “soft-power” strategies by which entrepre-neurs compete in highly ambiguous markets. Overall, we develop a holistic view oforganizational boundaries and offer insights into institutional entrepreneurship andresource dependence theories. Our most important contribution is reinvigorating thestudy of interorganizational power.
Organizational boundaries are fundamental. Ev-ery organization needs to establish its boundary todistinguish the organization from the environmentand define its domain of action (Aldrich & Ruef,2006; Scott, 2003). Given this central role, the phe-nomenon of organizational boundaries has beenaddressed with a set of rich theoretical perspectives(Santos & Eisenhardt, 2005). A first line of researchadopts an exchange efficiency view, looking at costminimization as a key driver of boundaries (Dyer,1996; Williamson, 1981, 1991). In a second stream,organizational boundaries are examined through apower lens, with a focus on how organizations cancontrol their exchange relations (Pfeffer & Salancik,1978; Thompson, 1967). Other research relies on acompetence view, in which the evolving resources
and capabilities of organizations shape theirboundaries (Brusoni, Prencipe, & Pavitt, 2001; Pen-rose, 1959; Peteraf, 1993). A fourth line of researchadopts an identity perspective, focusing on the cog-nitive frames of organization members that define“who we are” as an organization and shape thechoice of boundaries (Dutton & Dukerich, 1991;Porac, Thomas, Wilson, Paton, & Kanfer, 1995;Tripsas, 2009).
Yet, in contrast to the richness of these theoreti-cal perspectives, the range of research designs forstudying organizational boundaries has been nar-row. Indeed, much empirical research representsan atomistic view that focuses primarily on theantecedents of single-boundary decisions in cross-sectional samples (David & Han, 2004). The originof the atomistic view lies primarily in the earlyinfluence of exchange efficiency conceptions ofboundaries, notably transaction cost economics(TCE), for which the canonical problem is whetherto internalize or outsource a specific transaction(Williamson, 1985). This problem formulation hasled to the dominant approach of analyzing organi-zational boundaries as discrete structural alterna-tives such as make or buy. As a result, much em-pirical research examines boundary decisions inisolation, as if the shaping of organizational bound-aries were simply the accumulation of independentboundary decisions in well-structured settings. Al-though research has moved from efficiency-basedtheoretical explanations to include other decisiondrivers, such as competencies (Argyres, 1996; Jaco-bides & Hitt, 2005; Poppo & Zenger, 1998), theearlier mind-set of focusing on independent bound-
We gratefully acknowledge the research support pro-vided by the Stanford Technology Ventures Program andINSEAD R&D. We appreciate the thoughtful comments ofCharles Galunic, Morten Hansen, Sarah Kaplan, RiittaKatila, Woody Powell, Ray Levitt, Rory McDonald,Huggy Rao, Kaye Schoonhoven, Wes Sine, Bob Sutton,Patricia Thornton, and seminar participants at StanfordUniversity, INSEAD, the University of Colorado, the Uni-versity of California, Berkeley, Harvard University, theUniversity of Maryland, and Oxford University; and atthe Queen’s University Alliance Edge Conference, PrinceBertil Symposium, West Coast Research Symposium onTechnology Entrepreneurship, Atlanta Competitive Ad-vantage Conference, and inaugural LBS Entrepreneur-ship Conference. We also acknowledge the valuable con-tributions of Sara Rynes-Weller and three anonymousAMJ reviewers.
� Academy of Management Journal2009, Vol. 52, No. 4, 643–671.
643
Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download or email articles for individual use only.
ary decisions at specific stages in the “value chain”has largely prevailed.
Although understanding single-boundary deci-sions is valuable, this emphasis neglects potentialrelationships among decisions, ignores the inter-play of different boundary-setting mechanisms(e.g., alliances with key partners, identity position-ing, acquisitions), and does not address the evolu-tion of organizational boundaries (Santos & Eisen-hardt, 2005). Thus, it constrains understanding ofhow organizational actors actually conceptualizeand execute boundary work. In particular, this at-omistic emphasis leads to “losing sight of the forestfor the trees” (Jacobides & Hitt, 2005: 1209) byobscuring how individual boundary decisionsmight fit into overall patterns of strategic action.Thus, research designs that permit exploring theevolution of organizational boundaries and capturethe micro details of boundary work may reveal newinsights. Our aim is to address this research oppor-tunity by exploring the shaping of organizationalboundaries over time by new firms in nascent mar-kets. In this setting, the very beginning of organiza-tional life, boundary work is both crucial for sur-vival and poorly explained by current theories.
Nascent markets are business environments in anearly stage of formation, often appearing in emerg-ing “organizational fields” (Aldrich & Fiol, 1994).Nascent markets are characterized by undefined orfleeting industry structure (Eisenhardt, 1989a;Rindova & Fombrun, 2001), unclear or missingproduct definitions (Hargadon & Douglas, 2001),and lack of a dominant logic to guide actions(Kaplan & Tripsas, 2008; Porac, Ventresca, &Mishina, 2002). Thus, nascent markets constituteunstructured settings with extreme ambiguity. Fol-lowing organizational theory (Davis, Eisenhardt, &Bingham, in press; Weick, 1995), we define ambi-guity as lack of clarity about the meaning and im-plications of particular events or situations. Ambi-guity arises from unknown cause-effect relationsand lack of recurrent, institutionalized patterns ofrelations and actions (Aldrich & Fiol, 1994). Ambi-guity thus leads to confusion and multiple poten-tial interpretations. It differs from uncertainty,which refers to inability to predict the probabilityof specific outcomes (Davis et al., in press; Weick,1995), a situation that current boundary theories,especially those based on efficiency and resourcedependence logics, can deal with well.
Nascent markets are an intriguing setting inwhich to explore organizational boundaries be-cause these markets pose unique problems for or-ganizational actors. Executives operating in nas-cent markets typically lack a clear view of industrystructure; they may not know, for instance, which
organizations are their best prospects as customers,partners, competitors, and suppliers (Rindova &Fombrun, 2001). Thus, they are often unable tospecify cost functions (Gilbert, 2005), primary de-pendence relationships (Rao, 1994), competenciesthat are strategically valuable within the industry(Bingham, Eisenhardt, & Davis, 2009), and legiti-mated industry logics to guide action (Aldrich &Fiol, 1994; Kaplan & Murray, forthcoming). Suchmarkets both enable and reward strategic action byorganizational actors (Ozcan & Eisenhardt, 2009),yet it is unclear how existing theories of organiza-tional boundaries might apply in nascent marketswhere the basic elements of industry structure areambiguous, evanescent, or nonexistent.
New firms compound the problems of nascentmarkets. In contrast to established firms, youngventures typically have incipient activities and re-sources (Burton & Beckman, 2007; Rindova &Kotha, 2001), a fluid or nonexistent identity(Lounsbury & Glynn, 2001; Rindova & Kotha,2001), and little power to influence other firms(Hallen, 2009; Ozcan & Eisenhardt, 2008). Also,they face major strategic hurdles that make sur-vival, not efficiency, crucial (Graebner, 2004).These characteristics imply a set of firm attributesthat are not well addressed by existing boundarytheories, with their focus on established firms withsubstantial resources. Yet, at the same time, it islikely that the vulnerability of new firms makesorganizational boundaries pivotal. Given these ob-servations, the research setting of new firms in nas-cent markets seems likely to provide an excellentopportunity to extend current theories and revealnovel insights.
Scholars have begun to study boundary-relatedissues for new firms in nascent markets. One themecenters on cognitive and sociopolitical legitimacy(Aldrich & Fiol, 1994; Rao, 1994). Although muchof this work explores how institutional actors actcollectively to legitimate an emerging set of activi-ties (Leblebici, Salancik, Copay, & King, 1991;Rindova & Fombrun, 2001; Sine & David, 2003), afew studies address how entrepreneurs organizetheir own firms. For example, Rindova and Kotha(2001) examined how Yahoo executives used iden-tity to shape their actions in the nascent Internetsearch market. Similarly, Hargadon and Douglas(2001) studied Edison’s tactics for establishing thenascent electricity industry by framing it as cogni-tively proximate to the established gas lighting in-dustry. Although useful, these and other studies ofcognitive processes in ambiguous environmentsneither focus on boundaries per se nor address therich set of possible boundary theories and availablemechanisms (Santos & Eisenhardt, 2005).
644 AugustAcademy of Management Journal
A second theme is interfirm relationships (Pow-ell, Koput, & Smith-Doerr, 1996). Much of this workcenters on single-boundary decisions using re-source dependence and social network logics(Eisenhardt & Schoonhoven, 1996; Gulati & Hig-gins, 2003), but a few studies examine entrepre-neurs as they form holistic patterns of ties. Forexample, Powell and colleagues (1996) observedthat biotechnology ventures that form many tiesearly on gain valuable resources and attain morecentral network positions. Similarly, Ozcan andEisenhardt (2009) found that entrepreneurs whoengage in a strategy of forming multiple ties simul-taneously create more successful alliance portfo-lios. Hallen (2009) observed several alternativestrategies for forming successive ties in the garner-ing of resources from venture investors. But like theresearch on cognitive and sociopolitical legitimacy,these efforts neither focus directly on boundariesnor include a rich set of boundary mechanisms andtheoretical logics.
Overall, extant boundary research primarily ex-amines atomistic boundary choices in well-struc-tured organizations and environments. A few stud-ies take a longitudinal view and focus on emergingfields, yet they typically use a single theoreticallens and do not address boundary setting per se.Thus, there is a superb opportunity to develop amore complete and theoretically rich understand-ing of boundaries. We seek to extend current theoryand create new insights by studying the boundary-shaping processes of new firms in nascent markets.We ask, How do entrepreneurs addressing nascentmarkets shape their organizational boundariesover time?
Our research design is a multiple-case, inductivestudy that uses in-depth archival and field data totrack closely how entrepreneurs at five new firmsin different nascent markets shaped their organiza-tional boundaries during the initial years of organ-izational life. This design uses a research setting inwhich organizational boundaries are crucial, un-derexplored, and underserved by current theories.It also enables a longitudinal view and addresses abroad range of boundary mechanisms and theories,thus ensuring that we capture the richness andinterrelationships of boundary work. As such, ourresearch is consistent with recent calls for strategicand longitudinal studies of boundaries (Jacobides &Billinger, 2006; Santos & Eisenhardt, 2005).
Our central theoretical contribution is a holisticframework of the longitudinal processes by whichsuccessful entrepreneurs shape their organizationalboundaries and construct new markets. We findthat actors rely on multiple boundary mechanismscentered on three processes: claiming a market,
demarcating the market, and controlling it. Theseprocesses generate cognitive (identity-based), rela-tional (alliance-focused), and resource (acquisition-driven) structures for firm boundaries and nascentmarkets. We contribute to institutional entrepre-neurship by emphasizing novelty and dominance,not just fitting in and legitimacy. We also extendresource dependence theory to ambiguous environ-ments and entrepreneurial actors. Our key insightis that power is the unifying boundary logic. Bypower, we mean the ability of an actor to influencethe behavior of others in ways that produce out-comes favored by the focal actor (Pfeffer & Salancik,1978). In particular, we highlight how entrepre-neurs use soft-power strategies based on subtle per-suasion to dominate new markets, rather than tra-ditional hard-power tactics of coercion based onextensive resource control (Nye, 2004). Our keycontribution is thus reinvigorating the study ofinterorganizational power and reaffirming thatagency and strategic action often rest on the ration-ale of power, however subtly exercised.
METHODS
Research Design and Setting
We used an inductive, multiple-case research de-sign (Eisenhardt, 1989b). Multiple cases permit areplication logic in which cases are treated as ex-periments, with each serving to confirm or discon-firm inferences drawn from the others (Yin, 1994).This process typically yields more robust, general-izable theory than single cases (Eisenhardt & Graeb-ner, 2007). Our design embeds three units of anal-ysis: boundary decision, organization, and market.
The research setting is the confluence of comput-ing, electronics, and telecom industries in the midnineties. At this time, many innovations, such asdistributed computing, electronic messaging, andInternet commerce, began to gain widespread ac-ceptance. This setting was attractive because of theemergence of numerous nascent markets in thisindustry confluence and the related burst of entre-preneurial foundings. Our focus is the evolution ofthe organizational boundaries of five new firms.These firms were selected in mid 2000 from thepopulation of U.S. firms founded in late 1994 andearly 1995. This timing was attractive because itcorresponded to the period of highest ambiguity—that is, just prior to the take-off of the Internet-related sector (Sine, Mitsuhashi, & Kirsch, 2006).This period was sufficiently distant to allow longi-tudinal patterns to emerge and yet sufficiently re-cent to allow accurate, detailed data collection atthe time of our study.
2009 645Santos and Eisenhardt
Having defined the study’s population, we thencreated a diverse sample. We selected firms ad-dressing five distinct types of nascent markets: vir-tual marketplace, digital services, online com-merce, distributed enterprise software, andnetworking hardware. We also selected firms withdistinct founding contexts, founding teams, andinitial funding. The entrepreneurs ranged from alone engineer who stumbled serendipitously intoan opportunity, to entrepreneurs who owned atechnology and were searching for a market oppor-tunity, to seasoned executives with detailed busi-ness plans and strong ties to professional investors.Table 1 summarizes the diverse characteristics ofthe sampled firms. Studying such a diverse set offirms offered firmer grounding of theory thanstudying a more homogeneous one (Harris & Sut-ton, 1986).
Given our aim of understanding how organiza-tional actors shape boundaries, we employed a lon-gitudinal design that comprehensively tracks mul-tiple boundary decisions during the firms’ initial
five years. This design required that we study firmswith rich archival histories and willingness to grantdiverse, multiple interviews. These criteria furthernarrowed our choice to the five firms that we se-lected. Although our firms may be longer-livedthan many new firms, a sufficient history for eachfirm was necessary to understand the temporal dy-namics of setting boundaries. This requirementoutweighed having a random sample, especially ina process-focused and theory-building study suchas ours (Siggelkow, 2007). Similarly, although ourfirms were more successful than most new firms,they nonetheless exhibited much variation in theirboundary decisions and outcomes, such as failedactions and major mistakes, bringing useful vari-ance to our theory building.
Data Collection
We focused data collection on tracking theboundary decisions of each firm during its first fiveyears. In keeping with our interest in a rich and
External informants Industry expert Industry expert Industry expert Industry expert Industry expertCompetitor Partner Competitor Competitor CompetitorEx-employee Ex-employee Ex-employee PartnerPartner
a All firms were founded between late 1994 and mid 1995 in the United States and eventually went public. The names used here arepseudonyms.
646 AugustAcademy of Management Journal
longitudinal understanding of organizational bound-aries, we broadly defined a boundary decision as anorganizational choice that shapes the demarcationof an organization relative to its environment (San-tos & Eisenhardt, 2005; Scott, 2003). This broaddefinition allowed us to focus our inquiry on shap-ing boundaries while avoiding a restrictive theoret-ical or empirical lens. For example, this expansiveapproach includes boundaries specified by a firm’sresource portfolio (Brusoni et al., 2001), sphere ofinfluence (Pfeffer & Salancik, 1978), organizationmembers’ cognitive mind-set (Tripsas & Gavetti,2000), and governance of activities (Williamson,1991). Examples of boundary decisions are acquir-ing a firm, ending an alliance, redefining organiza-tional identity, and making an outsourcing choice.
We relied on two primary data sources: archivesand interviews. We began data collection by gath-ering extensive archival data from both internaland external sources. The internal sources in-cluded all press releases since firm founding (about50 per year per firm), Securities and ExchangeCommission (SEC) filings and initial public offer-ing (IPO) prospectuses (approximately 1,000–1,200pages per firm), internal reports and presentations(about 70–150 pages per firm), as well as video andaudio archives of presentations made by firm exec-utives at various points in time (an average of fourper firm). The external sources included media ar-ticles about each firm identified using ABI Inform.We located about 80 to 200 articles per firm, usingthe firm name as a keyword. We complementedthese sources with analyst reports, books abouteach firm when available, and media articles aboutcompetitors and the relevant nascent markets. Us-ing these very extensive archival data, we devel-oped chronological case histories for each firm.Each case was about 60 pages long and took abouttwo months to write. We began by developing achronological list of boundary decisions for eachfirm and then used these decisions to structure thecase histories. We organized the cases by year, de-tailing the relevant boundary decisions. We devel-oped tables and graphs for each case, includingtracking key metrics (e.g., revenue, market share,employees, and growth rates) and changes in thecomposition of the executive team (Miles & Huber-man, 1994).
We continued data collection using a second pri-mary source: semistructured interviews with inter-nal and external informants. We conducted an av-erage of 9 interviews per firm, accumulating a totalof 46 interviews from early 2001 to mid 2002. Ourfirst interview was typically with the CEO and/orfounder and lasted several hours. We used thisinterview to identify major boundary decisions,
which we then matched with those identified inthe archival material, thus triangulating the data.We defined a boundary decision for our informantsusing our formal definition (given above) andadded conceptually consistent lay language, de-scribing such decisions as “choices that shape firmscope and its domain of action.” These first inter-views were very helpful in validating the initial listof boundary decisions. A key advantage was iden-tifying decisions that were unavailable in archivalsources (e.g., major acquisitions not consummated,alternative identities not chosen, and outsourcingopportunities not pursued). After these initial in-terviews, we identified 13 to 15 major boundarydecisions for each firm and at least three internalinformants who could provide firsthand accountsof how each decision evolved.
We based selection of internal informants onthree criteria: (1) long tenure in their firm, whichwould provide a temporal perspective on the firm’sboundaries; (2) direct involvement in at least somemajor boundary decisions, which would providedeep, first-hand knowledge; and (3) functional andhierarchical variety, which allowed us to obtain avariety of perspectives. We complemented theseinternal informants with four types of external in-formants: former employees, business partners,competitors, and industry experts. Use of multipleinformants mitigates the potential biases of anyindividual respondent by allowing information tobe confirmed by several sources (Golden, 1992;Miller, Cardinal, & Glick, 1997). Use of multipleinformants also enables inducing richer and moreelaborated models because different individualstypically focus on complementary aspects of majordecisions (Dougherty, 1990; Schwenk, 1985).
The interviews ranged from 45 minutes to twohours in length. We recorded and transcribed them,generating about 800 double-spaced pages. The in-terview guide had two main sections. The first sec-tion was composed of open-ended questions thatenabled the informants to provide a broad view ofthe evolution of the relevant nascent market, thefocal firm, and its boundaries. The second sectionfocused on specific boundary decisions in whichthe informant was directly involved. We asked in-formants to relate the chronological story of thedecision as they observed it, prompted by probingquestions from the interviewer. At this stage thequestions concentrated on facts, events, and directinterpretations, rather than hearsay or vague com-mentary (Eisenhardt, 1989b). We further reducedthe potential for retrospective bias by triangulatingdata, matching real-time archival data with the ret-rospective accounts. Our epistemological approachwas thus to understand the meaning making and
2009 647Santos and Eisenhardt
conceptualizations of informants, while ensuringthat those interpretations had substantively in-formed behaviors and were not a product of laterimpression management.
The informants usually agreed about the factsand events of a given boundary decision. But theyoften also revealed complementary information.This allowed reconstruction of the histories ofboundary decisions in rich detail from variousviewpoints. Often informants described extensiveconflicts related to given boundary decisions.These accounts of conflicts were invaluable for ourexploring the roles of different boundary concep-tions in shaping boundaries. A key strength of ourdesign is that these interviews revealed details andmotives for decisions that were unavailable in thearchival data. Another strength is the use of exter-nal informants who offered an outsider perspectiveon boundary decisions and brought a “realitycheck” to the internal accounts.
Over six months, we then blended the interviewdata into the archive-based cases. The final casechronologies are about 100 pages long for eachfirm. We analyzed the context, decision process,implementation, and outcome of each boundarydecision and assessed its impact on firm and mar-ket boundaries. Overall, the combination of archi-val sources and interview data from internal andexternal informants enabled a rich, triangulated,and relatively accurate understanding of the phe-nomena (Kumar, Stern, & Anderson, 1993). For ex-ample, media articles identified key organizationalevents and clarified the industry context affectingeach boundary decision, while interview data re-vealed debates within the firms, hidden intents,and decision processes. The result is a relativelycomplete, robust understanding of the emergenceand evolution of boundaries.
Analysis
We began with an in-depth analysis of each casethrough the lens of our research question (Eisen-hardt, 1989b): How do entrepreneurs addressingnascent markets shape their organizational bound-aries over time? We had no theoretical preferencesor a priori hypotheses. We read the cases indepen-dently to form our own views of each case. The goalwas to identify independently the theoretical con-structs, relationships, and longitudinal patternswithin each case and with respect to our researchquestion. We used tables and graphs to facilitateanalyses (Miles & Huberman, 1994). We each de-veloped an understanding of the major boundarydecisions, which we reconciled by going back tothe data and, occasionally, back to the informants.
We also identified interactions among boundarydecisions and found connections among emergingcategories, which led to the specific patterns ofdecisions that emerged from the data.
We then turned to cross-case analysis, in whichthe insights that emerged from each case were com-pared with those from other cases to identify con-sistent patterns and themes (Eisenhardt & Graeb-ner, 2007). Focal firms and decisions were groupedrandomly and by variables of potential interest tofacilitate comparisons and develop propositions.Comparisons were initially made between variedpairs of cases. As patterns emerged, other caseswere added to develop more robust theoretical con-cepts and causal relations. Discrepancies andagreements in the emergent theory were noted andinvestigated further by revisiting the data. We fol-lowed an iterative process of cycling among theory,data, and literature to refine our findings, relatethem to existing theories, and clarify our contribu-tions. The data analyses took another six monthsand resulted in a theoretical model of how entre-preneurs shape boundaries in nascent markets.
Our data suggest that successful entrepreneurs innascent markets adopt a strategic approach to shap-ing organizational boundaries. For them, settingboundaries is central to the challenge of succeedingin highly ambiguous, competitive markets. We findthat, to address this challenge, entrepreneurs inter-twine organizational boundaries with market con-struction to achieve market leadership and a defen-sible position. That is, they adopt an almostmonopolistic imperative of dominating a distinctmarket that they construct. As a CEO explained,“These are the times that new technology fran-chises are being built and the way you win is thatyou have to be #1 by a long shot.”
Our data indicate that entrepreneurs enact thismonopolistic imperative using patterns of interre-lated boundary decisions that are organized intothree processes: claiming, demarcating, and con-trolling. That is, they attempt to: (1) claim a newand distinct market space and become its “cogni-tive referent” through identity-based actions; (2)demarcate this market by specifying firm and mar-ket boundaries through alliances with establishedfirms; and (3) control the market by overlapping theboundaries of the firm and market over timethrough acquisitions that eliminate entrepreneurialrivals. Underlying these processes is the unex-pected use of power tactics, such as creating illu-sions, using strategic timing, and exploiting the
648 AugustAcademy of Management Journal
tendencies of others, that form the strategic arsenalof entrepreneurs in nascent markets. Together,these processes enable a new firm with limitedinitial resources and potentially high dependenceon established firms to construct a distinct marketand achieve dominance in it. We develop eachprocess in detail next.
Claiming the Market
Entrepreneurs in nascent markets face an ambig-uous environment with unclear customers, unde-fined product attributes, and no well-establishedindustry value chain. Our data suggest that, in lightof this ambiguity, entrepreneurs in nascent marketsoften devote significant effort to claiming the mar-ket—that is, defining a distinct identity for both thefirm and market so that the two become synony-mous. If entrepreneurs are successful, their firmbecomes the cognitive referent for the claimed mar-ket: the organization that relevant others (e.g., cus-tomers, partners, analysts, and employees) auto-matically recognize as epitomizing the nascentmarket. For example, Google is widely seen as thecognitive referent for Internet search.
Our data indicate that entrepreneurs use threeidentity mechanisms to claim a market: adopt tem-plates, signal leadership, and disseminate stories.“Adopt templates” is defined as using well-knowncognitive models from other domains (either in iso-lation or combination) to convey a unique identity.This identity simultaneously makes a firm and itsnascent market distinct, yet also familiar and un-derstandable to market audiences. “Disseminatestories” is defined as spreading symbolic narratives(real or fictitious) to raise awareness about the firmand its market, and communicate the firm’s iden-tity. “Signal leadership” is defined as taking con-crete actions that convey superior power and ex-pertise within the market.
A good example is Secret, one of the firms westudied. Secret’s founders began with a sophisti-cated cryptography technology, but without a clearidentity or a well-defined product or customer set.They experimented with several ideas before gain-ing traction with an unexpected service: providinga secure environment for digital communications.But although this idea seemed promising, it wasalso very ambiguous. As the venture capitalist (VC)who backed the firm put it,
There was a product out there but it was not verywell-defined and it was searching for a market.
A Secret executive concurred:
At the time, it was the wild wild west—there was noplaybook for the Internet or our space—we createdthe playbook.
Secret’s executives spent considerable time try-ing to hone this idea by grappling with questionssuch as: What are we selling? Who are we? Who’sthe customer? In particular, they debated “secu-rity” versus “trust” as the core element of theiridentity. One executive describes the decision infavor of trust, which was an unusual identity in thisnascent market:
We believed that we had a broader obligation to theInternet at that time, which was to have this under-lying trust infrastructure. . . . Trust was not just se-curity in terms of keeping people out but it also wasletting people in. And we realized that a lot of whatwe did—digital certificates, digital signatures, thatwas not really security technology. . . . It was a trusttechnology.
Though Secret’s executives designated trust ascentral to the firm’s identity and conception of themarket, ambiguity remained. For example, one ex-ecutive noted, “It was a trust technology” but thenwent on to ask: “Is it a trusted service? Is it trustservices? Is it trust in infrastructure services?” AsSecret executives struggled to define their identity,they began adopting templates from seemingly dis-tant but cognitively related areas to describe theiractivities for would-be customers, other stakehold-ers, and even themselves. They used well-knownterms such as “ID card,” “wallet,” and “passport”as part of their vocabulary. Secret’s VC backer de-scribed how they explained their market: “Youknow, you have kind of an electronic wallet andhave all your IDs on one thing, and it would be-come your passport around the Net.”
To further clarify their identity and win accep-tance as the cognitive referent of the nascent mar-ket, Secret executives also began to signal leader-ship. For example, they hired a high-profile lawyerto convert the venture’s emerging operating proce-dures into a “best practices” framework that waspromoted at numerous venues. This framework be-came the market standard, serving to further clarifythe meaning of the nascent market to Secret’s ad-vantage and bolster the perception of the venture asthe market expert. Secret’s CEO explained:
One of the most significant early employees wasGeorge. He was very well thought of in both theacademic as well as the legal community. Georgespearheaded all our efforts . . . on encryption, digitalsignature law, what we call certificate authoritypractices. And we created the first set of industrypractices. The policies by which you should hirepeople, the policies by which we issue a certificate,
2009 649Santos and Eisenhardt
the policies by which we should revoke it. . . . So weinvented this as we went along and George’s abilityto put legal underpinnings to it really separated usfrom the would-be competitors that started in thegarage with a website.
Secret’s executives also relied on disseminatingstories that differentiated Secret and conveyed itsunique “trust identity.” For example, they orga-nized elaborate ceremonies for opening data cen-ters and invited the media (note: Secret operatedgeographically distributed data centers to deliverits service). These ceremonies were designed totransmit the image of trust through features such asarmed personnel and bunker-like facilities. Secretexecutives brought together players from the offline(e.g., notaries) and online (e.g., network securityexecutives) “trust worlds” to attend the ceremo-nies. Given the novelty of these ceremonies, Secretsucceeded in enticing the media into covering theventure in detail. For example, one reporterobserved:
An unusual ceremony at the new bunker-like oper-ations facilities of Secret grabbed the attention ofcertificate authorities such as notaries and accoun-tants, as well as corporate and network securityexecutives. Complexity and importance were ele-ments of Secret’s events. Witnessing the ceremonywere representatives from various organizations,armed officers, and a notary-videographer to recordthe ceremony for archiving.
A year later, Secret executives adjusted theiridentity by adding the template of a “public utility”that conveyed the ubiquity and high reliability of atrusted service. This identity guided later boundarydecisions such as which activities to pursue. Theyshunned even profitable activities that rivals ac-tively pursued if those activities fell outside Se-cret’s identity boundary. An example is the bound-ary decision to not engage in software sales. As aSecret vice president (VP) explained:
We decided that strategically we were a servicescompany. . . . We told a whole bunch of people:“Here is why the services model works, here is whyit is great, here is why we’re a services company todo these types of things. . . .” So you would notdecide: “Well, we are selling some software aswell.” You have to be consistent and that gives youcredibility. A key part there is that we did not reallywaffle. There were always people in sales or otherparts of the company saying: “Hey we can sell soft-ware too, it’s easy to sell, the customer can touch it,you get the revenue recognition in the current quar-ter for it, etc.” We said, “No, we are a service com-pany, we are staying on course here and stay in thisservices space.”
Over time, Secret executives succeeded in mak-ing the venture’s identity synonymous with thenascent market. By combining a trust vocabularywith the public utility template, disseminating sto-ries through symbolic ceremonies and press re-leases, and signaling leadership by setting onlinecertification standards for the market, the entrepre-neurs both clarified their venture’s identity andintertwined it with their conception of a market fortrust services (a conception that was distinct fromcompeting market conceptions such as selling se-curity products). This reinforcing pattern of bound-ary actions helped Secret to become the cognitivereferent for the nascent online certification market.Media reports from the period confirm this, callingSecret “the leading authority for certifying Webserver encryption keys.”
Magic, another sample firm, is a second example.Secret’s identity was initially poorly defined, butMagic began with a sharp sense of identity. Thecore of this identity was “customer-centric onlineshopping.” Nonetheless, Magic executives still hadto convince relevant others (e.g., customers andfinancial backers) that theirs was a winning iden-tity. Indeed, when Magic was founded, onlineshopping was a novel concept and one that wasdifficult to understand because it involved a verydifferent user experience than did traditional retailshopping. There was confusion even around basics,such as how to evaluate products and how to pay.Although the new technology offered striking fea-tures that were not available offline, Magic’s exec-utives nonetheless adopted templates from thebroad domain of offline shopping that were veryfamiliar to end users. For example, Magic’s userinterface was based on well-known, offline retail-ing concepts such as “shopping cart” and “checkout.” Their purpose was to reduce ambiguity andaccelerate user adoption.
Magic executives also signaled leadership bypurposefully offering the world’s widest selectionin their product category. They also located a fewfar-flung customers so that they could accuratelyclaim to be sending products to 45 countries and all50 U.S. states in their first month. They widelybroadcasted this achievement to the media, pro-claiming Magic to be the largest retailer in theworld in their category. But although this claimwas symbolically true in terms of geographic andproduct breadth, actual revenue was miniscule.Magic executives continued signaling leadershipby launching ads featuring high-profile figuresfrom academia, business, and sports comfortablyusing Magic’s offering. These actions supportedMagic’s identity as “customer-centric” and the ven-
650 AugustAcademy of Management Journal
ture’s positioning as the leading company in thenascent market.
Magic was also active in disseminating storiesthat reinforced its identity. For example, executivesreleased stories about the founder’s personal pas-sion for the customer (for instance, a story about acustomer with an unusual need that the founderinsisted that Magic satisfy) and corporate frugality(e.g., cheap office furniture) that reinforced theidentity of favoring customers over employeeperks. They also widely disseminated another storythat used a vocabulary that drew an analogy to U.S.history, portraying the founder as a “pioneer mov-ing west” to open up “a new frontier.” This storyfurther emphasized the venture as familiar, yetnovel.
This combination of identity mechanisms in-creased the likelihood of Magic becoming the cog-nitive referent for a distinct nascent market. In fact,media and analyst reports from the time stronglysupport the view that Magic became the cognitivereferent of its new market. As an outside expertconfirmed, “Magic has become the default namewhen you think of buying on the Net.” Moreover, itwas clear from the interviews and internal archivaldata that Magic executives purposefully craftedthese strategic actions to shape outsiders’ cognitiveunderstanding of their venture. As a senior execu-tive noted,
We knew that by the end of 2000, we would prettymuch have defined what the company stood for incustomers’ minds. . . . So you have to do that stuffpretty fast otherwise by the time you get around todo it you’re done! You can’t change people’s per-ceptions about what you are.
Although all firms engaged in practices aimed atclaiming a market, they were not equally effective.An example is Haven. Haven’s founder stumbledupon the nascent market of online marketplaceswhen his hobby unexpectedly became a success.He personally valued egalitarianism highly. Oneobserver described the founder as seeking “a fair,open, honest marketplace.” Another noted, “Havenwas capitalism for the rest of us.” To coalesce thesepersonal values into an organizational identity, thefounder adopted a “community” template, empha-sizing related vocabulary such as how “friends”could connect, share information, and trade in a“safe neighborhood.” Haven even installed “streetsigns” at the firm’s office to reinforce the commu-nity identity.
However, though this identity became very clearwithin Haven, it was unclear whether the identitywould become the cognitive referent. One chal-lenge was communicating this identity. An indus-
try expert recalled, “It was a totally different ani-mal; they [customers] didn’t know what to make ofit.” A second challenge was winning against doz-ens of competing market conceptions. For example,while Haven’s identity was a community of friendstrading with one another, Haven’s leading compet-itor (an older, larger firm) conceptualized its iden-tity as “Las Vegas style shopping excitement” andoffered a gambling format aimed at men. It wasunclear which (if any) of these very disparate iden-tities and related market conceptions would win inthe nascent market.
Haven’s founding team tried to gain traction inthe media with their identity by promoting a veryrational (and factual) account for the firm’s exis-tence based on a balanced, fair marketplace forbuyers and sellers. When this approach failed, afrustrated employee came up with the idea of dis-seminating a story about the company’s foundingthat illustrated Haven’s “community identity.” Un-like the rational account, this story had warm, per-sonal (albeit fictitious) details about the founderand illustrated how customers might use the ser-vice. This unusual and romantic story was pickedup by the press, embellished, and repeated in manyarticles featuring Haven. This media attention rein-forced the firm’s community identity and sharp-ened understanding of how customers could use itsservices. Haven executives followed up with mar-keting initiatives that further reinforced the coreidentity features of the story.
In addition, Haven executives then signaled lead-ership through preemptive litigation threats thatblocked potential competitors’ access to Haven’scustomers. Although they justified this hostile ac-tion in the media as protecting the “security” and“privacy” of the “community,” they were alsosignaling aggressive defense of Haven’s claim onthe market, despite its having lower resourcesthan some competitors. Overall, although Ha-ven’s late start in claiming the market probablyprecluded the firm from becoming the cognitivereferent in its first years, these persistent effortshelped to construct a distinct market and enabledHaven to be a leading firm within it. A mediareport confirmed that the “consumer-to-con-sumer online model is a new niche that Havenwas able to foster.”
Finally, as Saturn illustrates, entrepreneurs weresometimes unable to construct a distinct market inwhich they became the cognitive referent. Saturn’sfounders targeted an “empty space” located nearthe telecom equipment and networking markets.The founder explained:
2009 651Santos and Eisenhardt
Not one customer in the telecom business said theyneeded such an IP router for the core. Everyone withno exception was using Internet switches. Not onetelecom provider was even thinking of buildingsuch a router for the public telecommunicationsnetwork because no customer had ever askedfor one.
Although Saturn executives wanted to claim themarket, they did not use identity mechanisms well.For example, they did not adopt a template from adistant but cognitively related area. Rather, theyfollowed the template and vocabulary of the nearbynetworking market. They also kept their technologysecret early on, which impaired their ability to sig-nal leadership. Although they created a rationalefor the firm, it took the form of a “theory” to explainthe existence of their market. As such, it was sim-ilar to Haven’s failed initial rationale. Haven exec-utives, however, then created a memorable storywith compelling characteristics; Saturn executivesdid not. Moreover, they narrowly disseminatedtheir “theory” to industry analysts. Overall, lesseffective use of identity mechanisms limited Sat-urn’s ability to become the cognitive referent in adistinct nascent market. Saturn executives and in-vestors strongly believed that the firm was creatinga distinct market; for instance, its CEO argued, “Wehave a combination of a fortunate timing equationand a focused objective in what will become, whenhistory is written, a fundamentally different mar-ket.” Yet market audiences considered Saturn’smarket to be an extension of an existing market. Asa media report of the period noted, “Saturn iswidely perceived as a threat to [the market leader’s]hold on the networking market.”
Overall, our data indicate much variation in theuse of identity mechanisms to claim a market. Ta-ble 2 summarizes our data on claiming the market.More significantly, entrepreneurs who proactivelyuse reinforcing identity mechanisms (i.e., adopttemplates, disseminate stories, and signal leader-ship) are more likely to become the cognitive refer-ents in the distinct markets that they claim. Thus,claiming is a “sensegiving” process by which anentrepreneur can achieve cognitive dominance in anascent market.
The claiming process is effective for several rea-sons. First, adopting templates exploits the ten-dency of individuals to be attracted to the blend ofnovelty and familiarity (Davis, 1971). In keepingwith prior research, we observe that adoption offamiliar templates makes it easier for outsiders tounderstand a firm and its innovations (Hargadon &Douglas, 2001) and for insiders to grasp their firm’sidentity (Rindova & Kotha, 2001). But we also con-tribute the insight that having a distant template is
important as well. Adopting templates and relatedvocabulary from very proximate markets makes itless likely audiences will be intrigued and perceivea firm as having a distinct market and identity (e.g.,Saturn). In contrast, templates drawn from seem-ingly distant contexts, related by analogy, are morelikely to be understood and remembered (e.g., Se-cret). Adopting a template that combines the famil-iar and the distant thus enhances the likelihood ofacceptance and so developing a winning identityfor a distinct market.
Second, stories are effective because they exploita second tendency of individuals—that is, to over-value vivid stories. Research shows that stories areparticularly effective because they memorably con-vey information. Individuals are much more likely,for example, to remember the implications of sim-ple, emotionally charged stories than to rememberfacts and quantitative information (Heath, Bell, &Sternberg, 2001; Nisbett & Ross, 1980). Moreover,our research contributes insight into the story char-acteristics that are especially helpful for promotingunderstanding of a new firm or nascent market.When stories take the user perspective, portray afirm or its members in unusual situations, and con-tain intriguing personal information, they are morelikely to become sensegiving devices that are am-plified by the media. Interestingly, even fictionalstories are effective if then enacted and made partof the perceived reality (e.g., Haven).
Third, leadership signals are effective becausethey convey firm importance while also often beingrelatively inexpensive (e.g., Secret’s lawyer; Mag-ic’s few distant customers). As such, entrepreneurscreate illusions that they are more prominent,larger, and important than they actually are. Al-though Zott and Huy (2007) found that effectiveentrepreneurs use symbols to convey their quality,we add that symbols such as leadership signals canalso be illusory exaggerations that are nonethelesshighly influential for gaining the attention and sup-port of market audiences.
Overall, the claiming process is a reinforcingblend of sensegiving activities that define the iden-tity of a new firm as synonymous with the nascentmarket. As suggested by the research in “institu-tional entrepreneurship,” these types of identity-defining activities can enhance legitimacy (Harga-don & Douglas, 2001; Lounsbury & Glynn, 2001).Institutional entrepreneurship research empha-sizes creating legitimacy, whereby an organization-al form or institution becomes “desirable, proper,and appropriate” (Suchman, 1995: 574). We, on theother hand, emphasize creating cognitive domi-nance, whereby an individual firm constructs amarket and becomes its cognitive referent. By rely-
652 AugustAcademy of Management Journal
ing on soft-power strategies of nuanced influencebased on early timing, self-serving illusions, andexploitation of others’ tendencies, entrepreneurstransform an ambiguous, contested opportunityinto a winning claim.
Demarcating the Market
Although cognitive dominance is crucial, com-petitive dominance is also imperative. Even as en-trepreneurs begin establishing their identities and
TABLE 2Claiming the Market
OverallRatinga
Identity Mechanisms
ResultbAdopt Templates Signal Leadership Disseminate Stories
Definition Use of cognitive models fromother areas, together withvalues, practices andvocabulary.
Concrete actions that conveysuperior expertise and/orpower.
Spreading of symbolicnarratives about thecompany and/ormarket (fictitious orreal).
Rationale Sensegiving by analogy tohelp internal and externalactors understand ventureand market.
Create legitimacy andperception of leadershipin the eyes of internal andexternal actors.
“Magic has become the defaultname when you think ofbuying on the Net.”
Secret ������ ��Adopted “trusted services”
template and vocabulary.Added “public utility”
template.
��Defined and disseminated
legal framework of“certification bestpractices.”
��Organized unusual
ceremoniesw/armed guardsand bunkerfacilities to attractthe media.
Cognitive referent, distinctmarket
“The market [analysts andcustomers] . . . establishedSecret as leading authorityfor certifying Web serverencryption keys.”
Haven ���� ��Adopted “community”
template, vocabulary, andrelated social values.
�Reactive use of litigation
against would-be rivals.
�After early misstep,
widely publicized(and false) romanticstory about founder.
Not cognitive referent, distinctmarket
“The consumer-to-consumeronline auction model is anew niche that Harbor wasable to foster.”
Midway ���� �Late adoption of the template
of “operating system forthe Internet.”
��Acquired several high-
profile firms.
�Created new category
of end users, the“E-generation,” butdid not promote itmuch.
Not cognitive referent, distinctmarket
“The company has carved outa marketplace that hasproven crucial tocomputing.”
Saturn ��Did not adopt a distinct
template from anotherarea.
�Belatedly secured three
flagship clients.Kept technology secret for
two years.
�Created “theory” to
justify new market,but it was neitherwidelydisseminated normemorable.
Not cognitive referent, notdistinct market
“Saturn is widely perceived asa threat to [market leader’s]hold on the [existing]networking market.”
a To rate the use of identity mechanisms, we assigned each firm a score of “�” for use of a particular action. We assigned “��” if a firmwas particularly early and proactive in using this mechanism.
b We measured becoming the cognitive referent for a market by whether the firm was portrayed in the media as the market referencethree years after founding. We measured the existence of a distinct market by whether the market was described in the media as uniqueand independent of related markets.
2009 653Santos and Eisenhardt
claiming their nascent markets, they also face con-siderable ambiguity regarding key dependenciesand exchange partners. Moreover, they are usuallysurrounded by powerful established firms that maydefine the nascent market as part of their own ex-isting market or as an attractive new market to enter(Markides & Gerosky, 2005). It is often unclear howexecutives in these large, established firms mayultimately perceive their roles in a nascent market.If they choose to be competitors, they can become asignificant threat.
Anticipating this threat, our data suggest thatentrepreneurs in nascent markets attempt to avoidcompetition with powerful firms from nearby mar-kets by co-opting them with alliance mechanisms.These mechanisms create viable industry roles(e.g., supplier, complementer, buyer) for these po-tential competitors. In so doing, entrepreneurs de-lineate the perimeters of their firms and shape thesurrounding industry structure to gain clarity andsupport for their constructed market and avoidcompetition from powerful firms. It is this processthat we call “demarcating the market.”
Our data indicate three alliance mechanisms thatentrepreneurs use to entice powerful firms to ac-cept industry roles other than that of competitor:revenue-sharing agreement, equity investment, andantileader positioning. A “revenue-sharing agree-ment” is an alliance mechanism by which a partnerfirm benefits from a nascent market through distri-bution, advertising, or supplier contracts with afocal venture. Such agreements provide the partnerwith an industry role and revenues from the nas-cent market without directly participating. Thus,established firms have an incentive (and often alegal obligation) to support the new firm and nas-cent market.
“Equity investment” is an alliance mechanismby which entrepreneurs allow partner firms topurchase financial stakes in their ventures.Again, this enables these firms to benefit from thenascent market without participating directly.Thus, although entrepreneurs sacrifice owner-ship and may need to share private information,they also reduce the incentive of powerful firmsto become competitors.
“Antileader positioning” is an alliance mecha-nism used when there is a very strong firm domi-nating a proximate market. Entrepreneurs seekother established firms to join an alliance opposingthis leader. For such an invitation to be credible,the market claimed by the venture must be seen asattractive for the leader to enter. If that is the case,the other established firms will support the newfirm in order to distract and counter the leader’sdominance. The downside is that entrepreneurs
often alienate this dominant firm and are likely tobecome its rival.
Saturn is an excellent example. As noted earlier,Saturn began in a nascent market that was proxi-mate to the networking and telecom equipmentmarkets. Saturn anticipated competitive threatsfrom established firms in both markets. These firmsmight regard the nascent market either as an exten-sion of their own market or as an attractive marketto enter, especially if Saturn succeeded. Therefore,Saturn executives proactively attempted to definetheir organizational boundary and that of the nas-cent market by pursuing alliances with five estab-lished firms. Saturn’s enticement was twofold: Itoffered equity investment to give these firms a fi-nancial stake in the nascent market while leavingthem free to focus their time and resources else-where. Then, in addition to this important lure,Saturn was willing to be the antileader to thedominant firm. This willingness was key. Thepotential partners feared the leader and battledthis dominant firm in multiple markets (e.g., op-tical, landline, wireless). As Saturn’s CEO ex-plained: “All of them were worried about [theleader] and we were the anti-[leader], so it was achance for them to team up with somebody thatwas taking [the leader] on.” To the surprise ofSaturn executives, all five target partners agreedto ally. Two then agreed to equity investments.When the others found out, they clamored toinvest as well. These alliances were costly toSaturn because they required disclosure of pri-vate information and diluted ownership whenthe firm already had sufficient financial re-sources. However, the goal of demarcating a clearperimeter that clarified Saturn’s boundary rela-tive to threatening firms outweighed the draw-backs. Antileader positioning and equity invest-ment thus transformed five very large firms frompotential competitors into partners and furtherdefined the industry structure. An industry ex-pert noted:
They pulled together a beautiful deal on the strate-gic side that was tried to be copied by many othercompanies. This is difficult when you are dealingwith so many gorillas.
As the market grew, successful co-optation ofSaturn’s partners required continual attention andfurther incentives to keep them from becomingcompetitors. For example, a year later, Saturn ex-ecutives deepened their alliances with several part-ners via revenue-sharing agreements. These weredistribution ties that legally obligated the partnersto stay out of Saturn’s market and promote Saturn’sproducts. In return, these partners gained a share of
654 AugustAcademy of Management Journal
the products’ revenues. An industry expert com-mented on the success of one of these agreements:“Saturn’s partner was selling a remarkable amount,like a hundred million dollars, of their gear. Thatwas a great partnership.”
Overall, Saturn was successful in demarcatingthe market by using alliance mechanisms to createviable industry roles for potential competitors anddeter their market entry. Of course, Saturn did notdeter the market leader, which became a competi-tor. Still, the combination of a strong technology,growing awareness of Saturn by key market audi-ences, and the transformation of some potentialcompetitors into partners enhanced Saturn’s com-petitive strength, leading to a duopoly between Sat-urn and the leader. As an analyst noted, “The mar-ket became a two-horse race.”
Secret is a second example. As noted earlier, itsexecutives were constructing a nascent market fo-cused on trust services in digital commerce. Theyanticipated that several large firms in the banking,telecom, and software markets might become com-petitors, especially if Secret succeeded. Soon afterthe firm’s founding, Secret’s executives organizedseveral brainstorms to identify the firms most likelyto become competitors in order to try to deter theirentry. Secret’s CEO explained:
One of the things that separate us is that we arealways worried about who could take this awayfrom us and we tend to find a way to cooperatewhen there is a win-win scenario. Before they rec-ognize us and turn their attention to us, we find away for them to benefit from their associationwith us.
Secret’s executives chose equity investment astheir initial alliance mechanism to co-opt thesefirms. They spent a few months negotiating a roundof equity financing with several of them. Muchlike Saturn’s executives, those at Secret were notparticularly concerned with young firms thatwere perceived by others as their rivals. An ex-ecutive confirmed: “We were not afraid of our[entrepreneurial] competitors. We were alwaysafraid of the bigger people . . . whether it’s FirstData or IBM or Microsoft.” So, their strategy was toanticipate the future moves of these establishedfirms before their executives recognized the prom-ise of the nascent market. They hoped that an eq-uity stake in Secret would encourage these poten-tial competitors to focus elsewhere. As Secret’s VCexplained:
I wanted to get all people believing in what we weredoing and starting to sell the vision of certificate usefor identification. And I wanted to keep them out ofthe market. . . . Everybody was doing a hundred
things. But if we took this one off their plate, hope-fully they would become our partner.
A few months later, Secret announced an equityinvestment round that included several large po-tential competitors from proximate markets. Al-though these alliances provided attractive re-sources and prestige for Saturn, our interview andarchival data clearly reveal that Secret executiveswere strategically focused on deterring competitionand demarcating the nascent market. They wantedSecret’s position clarified and wanted these poten-tial competitors to take on supporting industryroles. Thus, their actions were designed to set theorganizational boundary, demarcate the marketperimeter, and define nonrival industry roles.The CEO explained: “We have kind of also cre-ated a demilitarized zone for ourselves. I thinkthat was very important.” Secret’s VC went fur-ther: “They [the equity investment alliances] didhave material consequences. They established usas the leader.”
Anticipating that the effectiveness of equity in-vestments to deter competition might wane if thenascent market grew, Secret executives proactivelycontinued to discourage competition by offeringrevenue-sharing agreements. For example, Secretsuccessfully sought an alliance with a leading po-tential competitor by redirecting some of Secret’sactivities so that this firm became a supplier, not acompetitor. Secret executives also created an affil-iates program to entice potential international com-petitors to sign revenue-sharing agreements. Secretgave these firms a portion of in-country revenuesand a complementer role in their geographic re-gions. According to the CEO, “The affiliate modelwas put out so that we could get into bed withphone companies.” As a result, Secret avoidedcompetition with another group of powerful es-tablished firms. Overall, Secret’s alliances de-terred competition. In fact, no established firmchose to compete directly with Secret. As oneindustry expert noted, “Secret has little directcompetition.”
Saturn and Secret executives anticipated com-petitive threats from established firms and proac-tively blunted them; other entrepreneurs were nei-ther as prescient nor as effective. Nonetheless, evenmodest attempts at demarcating often sharpenedorganizational boundaries, further defined markets,and delayed competition. Haven illustrates thesepoints. Haven executives began using alliances todeter competition and demarcate the firm’s marketseveral years after founding (in contrast, Secret andSaturn began much earlier). Because of this delay,the nascent market was less ambiguous and more
2009 655Santos and Eisenhardt
attractive for established firms. Haven executivesidentified three powerful established firms thatwere significant competitive threats because oftheir well-known aggressiveness and proximatemarkets. These firms possessed crucial resourceadvantages (including much larger customer bases,deeper pockets, and more reliable technologies).Although Haven’s executives approached all threefirms with revenue-sharing agreements, they weresuccessful with only one. Even that deal was pain-ful for Haven. It deeply divided the executive teambecause it required Haven to sign an agreement thatincluded an expensive advertising contract (the ini-tial down payment consumed half of Haven’s an-nual marketing budget) in exchange for exclusiveaccess to that firm’s customers and a “noncompete”agreement. On the one hand, the business develop-ment team argued that the deal was necessary togain customers and avoid competition. On theother hand, senior executives were loath to give upso much money since Haven was already gainingcustomers. The deciding factor was their belief thatblocking competition from this leading firm wasworth the price. As a board member put it: “Wewere paying that amount to prevent [the large po-tential competitor] from entering the business.” Asin the other ventures, a power-driven, monopolisticlogic was decisive in shaping organizational andmarket boundaries.
Simultaneously, Haven executives were also ne-gotiating alliances with the two other possible com-petitors. Unfortunately, executives at these firmsperceived Haven’s nascent market as a very attrac-tive extension of their own markets. After severalmonths of secret negotiations, they each offered toacquire Haven instead of allying. Haven executivesdeliberately prolonged the subsequent acquisitiontalks to gain more time to establish Haven in themarket. They had no intention of being acquired.Eventually the acquisition offers were withdrawn.The CEO of one of those firms described his finalnegotiating instructions: “OK, then let’s stop talk-ing to them because we really want to build ourown. Then we’ll go kill them!” These two firmsintensely competed against Haven for two years.Nonetheless, the alliance with the first firm and thenegotiation delays with the others gave Haven alead. This lead, coupled with fortuitous networkeffects and an increasingly compelling identity (seethe prior section), helped Haven win over thecompetition.
As in claiming, in demarcating the market oursample exhibits useful variation: not all the sam-pled entrepreneurs engaged in market demarcation.At Magic, for example, executives were aware ofestablished firms that might compete with their
venture, but they did not pursue alliances to detertheir entry. Instead, when several large firms pro-posed alliances, Magic’s executives rudely rejectedthem. A Magic VP noted, “We thought that thosecompanies were old-fashioned and that they couldnever really compete with us.” Rather, Magic in-vested in equity alliances with small ventures. Un-fortunately, two rejected suitors entered the marketand competed surprisingly well against Magic.They forced the company into price wars and highexpenditures that brought four years of losses. Iron-ically, not only did Magic fail to avoid strong com-petition from established firms—the venture alsolost most of its investment in its small allies whenmany of these firms failed.
Table 3 compares the firms’ use of alliancemechanisms to demarcate markets. As shown,entrepreneurs who proactively use a mix of alli-ance mechanisms (i.e., equity, revenue sharing,antileadership) to define their organizationalboundaries are likely to face less competition fromestablished firms and create a more delineated in-dustry structure around their markets. Thus,though claiming activities help entrepreneurs toachieve cognitive dominance, demarcating is apowerful co-opting process that helps firmsachieve competitive dominance. Demarcating alli-ances cancel or at least delay competition fromestablished firms, thus favorably shaping competi-tive dynamics.
The demarcating process is effective for severalreasons. First, alliances exploit the natural ten-dency of large firms to delay entry into nascentmarkets until these markets are well defined(Ozcan & Eisenhardt, 2009). Executives in estab-lished firms often believe (and rightly so) that theycan delay entry and still be effective competitors(Markides & Gerosky, 2005). Moreover, they oftenprefer to approach nascent markets by creating“real options” such as ties to new firms that pro-vide learning opportunities and potential stepping-stones to accelerated market entry.
Second, these alliances are also effective becausethey enable entrepreneurs to use self-serving illu-sions. For example, entrepreneurs were more effec-tive in demarcating their markets when they cre-ated the impression that they were open to beingacquired by their partners even when they werenot. Similarly, when alliance attempts failed, entre-preneurs sometimes pretended that they wanted tobe acquired when they were actually delaying theirwould-be buyers to postpone competition and in-crease their own strength.
Third, the demarcating process relies on timing.We observe that demarcating is most effectivewhen entrepreneurs anticipate threats early, pre-
656 AugustAcademy of Management Journal
emptively approach the established firms, and offersomething very significant (e.g., an equity stake,revenue sharing, and/or strategic value against aleading firm that often becomes an enemy) to entice
potential competitors to be partners. Moreover, ef-fective demarcating is a continuous process. Asnascent markets become successful, establishedfirms become more interested in market entry.
TABLE 3Demarcating the Market
0verallRatinga
Mechanisms for Co-opting Alliances
ResultbEquity InvestmentRevenue-Sharing
Agreement Antileader Positioning
Definition Sale of equity stake toestablished players innearby markets.
Contract that givespayments to players innearby markets foradvertising, supply ordistribution.
Frame venture as mainadversary of thedominantestablished firm in anearby market.
Rationale Deter entry by givingindirect participationin success of firm.
Deter entry by givingdirect participation insuccess of firm.
Deter entry of lesserfirms by confrontingleading firm forthem.
Secret ���� �� �� Low competitionRound of financing with
ten key potentialentrants. Structuredindustry roles.
Co-opted largest player innearby market.Created affiliatesprogram for potentialcarrier entrants.
Five major firms becamepartners. No marketentry by establishedplayers.
“Secret has little directcompetition.”
Saturn ���� �� � � Medium competitionRound of financing with
five potentialentrants. Did notapproach leader.
Co-opted three majorplayers withdistributionagreements.
Attracted five of themajor competitors ofthe industry leader.
One player co-opted long-term, others delayedentry. Leader entered.
“The market is now a two-horse race.”
Haven �� �� High competitionApproached three of the
largest players innearby markets. Co-opted largest withadvertising deal,delayed others.
One major player co-optedlong-term. Othersdelayed, and thenentered.
“Competitive intensitybetween the companiesis rising dramatically.”
Midway �� � � High competitionCo-opted two major
players withdistribution contracts.Only delayed thembecause gave no majorconcessions.
Attracted lesser firmswith goal of dilutingpower of the leadingplayer.
Limited early competition.Then much competition.
“The company is facingbrutal competition fromlarger, more establishedadversaries.”
Magic � � Highest competitionDid not approach
established players.Arrogant whenapproached by them.
Created affiliates programto co-opt small playersin nearby markets.
Powerful players entered.Intense price wars, highinvestment, and losses.
“In the midst of a furiousbattle between Magicand competitor forhegemony.”
a To rate use of mechanisms for co-opting alliances, we assigned each firm a “�” for each mechanism used and a “��” if a firm usedthe mechanism early and proactively.
b We measured level of competition by the strength of rivalry in a market four years after founding, using external archival sources.“Low” refers to no significant competition; “medium” refers to a duopoly; “high,” to an oligopoly; and “highest,” to an oligopoly in whichthe focal firm was forced into strong price wars.
2009 657Santos and Eisenhardt
Thus, co-optation incentives need to be enhanced.For example, Midway (a venture establishing a newsoftware market) was able to form demarcating al-liances with major firms, turning potential compet-itors into complementers and buyers. But Midwaydid not enhance the enticements for these firmsover time. As the market blossomed, their partnersbecame competitors.
Finally, the demarcating process relates to therole of alliances in theories ranging from transac-tion cost economics and resource dependence tothe resource-based view of the firm. Advocates ofthese theories often regard alliances as a means toaccess resources (Dyer & Singh, 1998; Gulati, 1998;Hallen, 2009; Katila, Rosenberger, & Eisenhardt,2008). Although our data confirm that gaining re-sources can be a rationale for alliances, it is notalways the driving force. Rather, alliances can beused to delay competition and define a favorableindustry structure by deterring the entry of strongpotential competitors, clarifying a firm’s bound-aries vis-a-vis others, and creating supporting rolesfor potential competitors as suppliers, comple-menters, and buyers. Thus, we assert that powerand co-optation, not just resources, are central towhy firms seek alliances.
Controlling the Market
Entrepreneurs in nascent markets typically faceentrepreneurial rivals, and these may have differ-ent resource combinations and alternative marketconceptions. Given high ambiguity, entrepreneursare usually unable to anticipate whether these en-trepreneurial rivals will outcompete them, be irrel-evant, or be acquired by established firms as step-ping-stones into the market. Given the possiblecompetitive threat from these rivals, our data indi-cate that entrepreneurs try to control the market byoverlapping their organizational boundary with themarket boundary in such a way that their firmoccupies as much of the market space as possible.This is achieved through acquisition (and oftendestruction) of the resources of entrepreneurialrivals.
Our data indicate that entrepreneurs use threetypes of acquisition mechanisms to control mar-kets: elimination of competing models, increasingcoverage, and blocking entry. “Elimination of com-peting models” is an acquisition mechanism aimedat destroying the resources of threatening rivals.These rivals usually have resources or businessmodels that could be superior or otherwise damag-ing to a focal entrepreneur’s control of a nascentmarket. After the acquisition, these resources aredestroyed or blended into the acquiring firm to
make its resource portfolio more robust. “Increas-ing coverage” is an acquisition mechanism that ex-pands an acquirer’s presence into emerging areas ofa nascent market (e.g., new geographical regions,new categories of users) so that the boundaries ofthe market and firm continue to be aligned as themarket expands. “Blocking entry” is an acquisitionmechanism aimed at removing possible stepping-stones into a market. The possibility that estab-lished firms could easily enter the market by ac-quiring these rivals is the main concern beingaddressed, not fear of the rivals per se. This type ofacquisition often concludes with disposal of theacquired resources.
An example is Midway. Its founders addressed anascent market related to “middleware” (softwarethat functions in the area between an operatingsystem and applications), and they adopted theidentity of “operating system of the Internet.” Asdiscussed earlier, Midway executives tried to allywith several established firms that were potentialcompetitors. However, they also faced entrepre-neurial rivals that were addressing roughly thesame nascent market. Therefore, they took severalsteps to control the market. First, they increasedcoverage of the market by acquiring several smallrivals in different geographic locations. This acqui-sition spurt enhanced Midway’s market coverageand greatly reduced the possibility of viable rivalsemerging in these locations. Later, Midway phasedout the acquired products and dispersed their re-sources, except sales and service staff.
Having increased their market control, Midwayexecutives next contacted their main rival. Thisfirm had a similar technology but a different distri-bution model (i.e., a third-party sales channel) thatwas not working well. Its sales were stagnant. Al-though it was clear that Midway was gainingagainst this rival and did not need its resources(Midway possessed similar technology, better mar-keting resources, and its own sales force), Midwayexecutives still wanted to make a blocking entryacquisition to prevent established firms from easilyentering the market by buying this rival. Since thetwo ventures owned the only software productsbased on a critical technology, this acquisitionwould also allow Midway to block access to keypatents. Indeed, Midway’s CEO acknowledged theintentional obstruction of two specific establishedfirms:
The primary reason to make this acquisition was tomake sure that neither “A” or “B” companies endedup with the technology. . . . Their products todayare better than the old stuff but they can’t go all theway [without this technology]. So it was a blockingmove on our part.
658 AugustAcademy of Management Journal
Midway executives phased out the acquiredbrand and product and transferred customers andservice people to their own software product. As aresult, Midway became the undisputed marketleader and significantly delayed the entry of thetwo threatening firms. These established firmsneeded an additional two years to develop technol-ogy that could circumvent Midway’s patents.
Later, several entrepreneurial rivals entered themarket with newer technologies. Faced with this newchallenge, Midway executives assessed whethereliminating competing models via acquisition wasworth the cost. Although doing so was expensive,they decided to acquire two ventures with threat-ening technical resources. These acquisitions gaveMidway executives time to upgrade their own tech-nology, using in-house resources plus resourcesfrom the acquired firms. Overall, Midway execu-tives enhanced their market control through aggres-sive acquisitions that gave the firm broad coverage,eliminated competing resources, and removedstepping-stones into the market. Despite the strongchallenge from the leader in a proximate marketand the breakdown of its demarcating alliancesover time (see the prior section), Midway’s proac-tive use of acquisitions to control the markethelped the firm to reach roughly 60 percent marketshare five years after founding.
Another example is Haven. Three years afterfounding, Haven was becoming a leading firm in itsmarket. But it was not the cognitive referent, hadallied with only one established potential compet-itor, and faced competitive challenges from twoother established firms (see prior sections). Simul-taneously, numerous new firms were vying forroughly the same market space. Thus, Haven exec-utives decided that controlling the market was cru-cial to their survival. They set up a “war room” intheir headquarters, complete with camouflage dec-orations and military items, where they would reg-ularly discuss the moves of their most importantrivals. This militaristic mind-set translated into ag-gressive acquisition of entrepreneurial rivals tocontrol the market. First, Haven executives ac-quired a rival that was second only to Haven innumber of users. This company was not a threat byitself, since it was much smaller than Haven. Butestablished firms could have easily used it as astepping-stone into the market because its technol-ogy and customer base would have been valuablefor a quick market entry. Haven’s press release be-nignly framed the acquisition as a synergistic moveto add valued resources: “We believe that [the ac-quisition] will ultimately bring a great deal of valueto our communities.” Yet internal informants re-vealed a strategic logic very different from that in
the firm’s press releases and media coverage. Forexample, an executive stated, “The move [the ac-quisition] was as much defensive as it was offen-sive.” Another expanded:
The real concern was the prospect that a larger com-pany might buy this venture and use it to launch acompeting service. These concerns were wellfounded. Haven did not know it, but a large playerwas making such an overture to the venture at thesame time.
Not surprisingly then, Haven gave the acquiredfirm little postacquisition attention and quicklyphased it out. Nevertheless, the acquisition blockedthe entry of the two major firms with which Havenhad failed to ally and forced them to enter moreslowly via organic growth.
A year later, an entrepreneurial rival appearedwith an alternative market conception based on anovel business model. When this firm began togrow, Haven executives decided to make an acqui-sition to eliminate a competing model, a model thatwould be particularly threatening if acquired by apowerful player. An executive at Haven said, “Wealso felt that there were some companies interestedin [the alternative firm] and that could be a way forthem to build the competing business model.” Af-ter the acquisition, Haven executives added someaspects of that business model to their main offer-ing and so enhanced their resistance to similarthreats in the future.
Shortly after this acquisition, Haven executivesrealized that their market was becoming globalwhen they saw several entrepreneurial rivals imi-tating their business model in Europe and Asia.These firms, they saw, might eventually becomelarge and control their local markets, or be acquiredby a powerful firm looking to compete with Haven.Given the strong network effects present in theirbusiness, timing was of the essence. Therefore, Ha-ven executives made several acquisitions that in-creased the size of the firm’s footprint in the ex-panding global market. In one country, theytransferred the customers to Haven’s services whilethey disposed of other acquired resources. In theothers, they left the acquisitions as stand-alonebusinesses. Overall, by using acquisitions to blockentry, eliminate competing models, and increasecoverage, Haven executives strengthened their con-trol of the nascent market and overcame their earlyfailures in becoming the cognitive referent and co-opting powerful players. With these moves, cou-pled with helpful network effects, they gained mar-ket control, taking an 80 percent market share fiveyears after founding.
Midway and Haven were particularly aggressive
2009 659Santos and Eisenhardt
in attempting to control their markets; the othersampled firms were less so. For example, after theirearly success in claiming the market, Magic’s exec-utives were too confident in their ability to defeatestablished firms. As a former employee recalled:
He [the founder] suffered from the classic problemof overconfidence . . . now, you need some of that ifyou want to be an entrepreneur and want to buildany level of success. You need unbridled ambitionand you need to be confident of yourself. What cantrip you up is if you’re overconfident.
So while Magic executives, led by their founder,considered acquisitions to block the entry of estab-lished firms, they did not pull the trigger. In onecrucial case, they decided that the price for a failingrival was too high. Unfortunately, an establishedfirm bought this rival, used its resources to entersuccessfully, cut prices, and gained market share atMagic’s expense. Although Magic executives madea few small acquisitions, these were aimed at inter-nalizing new technologies. Overall, Magic’s weakattempts to control the market were instrumental ineventually capturing only a 30 percent marketshare.
The most extreme case is Saturn, where execu-tives initially were philosophically opposed to ac-quisitions. Saturn’s CEO said, “I am very conserva-tive because I still think you build companies theold fashioned way and you cannot cheat [i.e., growby acquisition].” As noted earlier, Saturn had ini-tially developed strong alliance relationships withfive major established firms that helped to demar-cate the market. Although two of these firms re-mained partners, the others decided to competewith Saturn when they saw its success. Since Sat-urn had not attempted to acquire any of the newventures that had also sprouted in the market, theseventures were readily available as stepping-stonesfor Saturn’s erstwhile partners. The former partnersbought several of Saturn’s entrepreneurial rivals,fueled them with new resources, and quickly estab-lished market presence. One of these acquired ri-vals became a very significant competitor. Unfortu-nately, this occurred just as the leading firm (thedominant one against which Saturn had used anti-leader positioning) launched an aggressive attackagainst Saturn. Faced with the possibility of slip-ping into third place in the market, Saturn execu-tives finally began to use acquisitions to control themarket. But the price was high. For example, Sat-urn had to make a very rich cash-based offer of 30percent of Saturn’s market capitalization to buy afast-growing rival that other firms were also pursu-ing. Although late and expensive, this drastic moveenabled Saturn to maintain a market duopoly.
Overall, our data indicate much variation in theuse of acquisitions to control the market. Table 4presents these data. More significantly, entrepre-neurs who acquire entrepreneurial rivals to controla market are likely to achieve a higher market share.Thus, we identify controlling as an ownership-based process by which entrepreneurs favorablystructure competition. By using acquisitions toeliminate alternative resources, cover expandedmarket areas, and remove stepping-stones, entre-preneurs are able to reduce rivalry (a key feature ofindustry structure and predictor of profitability)and enhance competitive advantage by makingtheir own resources rarer (Peteraf, 1993).
The controlling process is effective for severalreasons. First, acquisitions exploit the tendency ofentrepreneurs to prefer acquisition by other ven-tures to acquisition by established firms. Indeed,research focusing on the seller side of acquisitions(Graebner & Eisenhardt, 2004) indicates that entre-preneurs prefer to be bought by other entrepreneurs(even at a lower price) because they perceivegreater cultural fit and a higher likelihood that theirventure will add value inside the buying firm. Al-though these sellers may later be disappointed(Graebner, 2009), this tendency to prefer entrepre-neurial buyers makes it easier for entrepreneurs toengage in controlling acquisitions because targetsare more likely to acquiesce and sell for a lowerprice, thus making them willing and affordablesellers.
Second, acquisitions are often effective becausethey enable entrepreneurs to engage in self-servingillusions. Since acquisitions are usually negotiatedin secret, entrepreneurs can more easily disguisetheir intentions. Graebner (2009) found that buyersoften create the illusion that they want to capitalizeon synergies and that they intend to “grow” theacquired firm. Indeed, our data indicate that publicstatements of entrepreneurs are often benign, buttheir private statements and later actions are clearlypredatory. Buyers often also emphasize the culturalfit between their firm and a target, although againthis may be illusory (Graebner, 2009). Such use ofillusions improves the chances that the target willagree to be acquired.
Finally, the controlling process has implicationsfor the resource-based view of the firm and theacquisitions literature. The resource-based viewemphasizes using acquisitions to obtain resources(Ahuja & Katila, 2001; Capron, Dussauge, & Mitch-ell, 1998) and achieve synergies (Graebner, 2004;Larsson & Finkelstein, 1999). Although we saw thislogic (e.g., Magic), we also observed that well-timedacquisitions were often used to cancel rivals andblock the entry of established firms. The destruc-
660 AugustAcademy of Management Journal
TABLE 4Controlling the Market
OverallRatinga
Controlling Acquisitions
ResultbElimination of Competing
Models Blocking Entry Increased Coverage
Definition andrationale
Acquisition of threateningrival firm due to differentor superior businessmodel or resources.
Acquisition of rival firm toremove a stepping stoneentry point for powerfulpotential entrants.
Acquisition of rival firm togain presence in newareas of the market (e.g.,new geographies, newcustomer types).
Outcome ofacquisition
Threatening businessmodels or resourcesclosed or combined.
Acquired firm closedalthough fungibleresources may be used.
Acquired firm carefullyintegrated to preserveresources/momentum.
Midway ������ �� �� �� LeadershipTwo acquisitions of
ventures with competingtechnology (one closed,one combined).
“We could potentially besideswiped by theemergence of thistechnology.”
One acquisition to controlall patents (closed).
“The primary reason wasthat neither A or Bended up with thetechnology.”
“Suddenly, we had aworldwide presence withpeople who knew how tosell and service.”
�60% share
Haven ����� �� �� �� LeadershipOne acquisition of
threatening businessmodel (combined).
“We could learn as muchfrom them as they couldlearn from us.”
One acquisition to blockentry by acquisitiveestablished firms(closed).
“The real concern was thata larger Internetcompany might buy firmand use it to launchcompeting service.”
Six acquisitions in bothnew types of customersand geographies(combined).
“We did not integrateimmediately becausethey were growing fast.”
�80% share
Secret ���� �� �� LeadershipOne acquisition of
competing technology(closed).
Four acquisitions to addnew types of customers(combined).
�80% share
“Their technology did nothave the samedependence as ours.”
“It was really just to giveus a beachhead in theenterprise market.”
Magic �� � � LeadershipOne small acquisition of
threatening businessmodel (closed down andrelaunched as newcomplementary service).
Failed to make a keyblocking acquisition thatbecame a stepping-stone.
Two acquisitions to addgeographic coverage.
“We bought companiesbecause it was a quickerway to get off theground.”
�30% share
Saturn �� � � Second placeOne late, very expensive
acquisition (combined).Failed to make key
blocking acquisitionsthat became stepping-stones.
Two late acquisitions toadd new customer types.
“What we wanted to dowas to get a footholdinto the cableoperators.”
�20% share
a To rate use of controlling acquisitions, we assigned each firm a “�” for each acquisition mechanism used and a “��” if a firm usedthat mechanism early and proactively.
b We measured market share by the percentage of total sales obtained by a focal firm in its market five years after founding andleadership by whether the firm had largest share. External archival sources were used for data.
2009 661Santos and Eisenhardt
tion of acquired resources is a particularly strikingfeature that we observed but that is largely ignoredin the resource-based literature. This finding sup-ports recent calls to examine decisions about re-sources using a power perspective (Chatain & Ca-pron, 2008). It also implies that the widely notedhigh failure rate of acquisitions may be overstatedbecause it neglects this resource-destroying ration-ale. In fact, many acquisitions in our data wereportrayed as failures in the media because they didnot lead to new business initiatives. Yet our dataclearly show that these acquisitions helped ven-tures to control their markets, making them actu-ally quite successful.
A MODEL OF ENTREPREURIAL ACTION INNASCENT FIELDS
In prior sections, we sketched the processes thatemerged from our data through which entrepre-neurs in nascent fields shape their organizationalboundaries and construct their markets. These pro-cesses suggest three corresponding propositions:
Proposition 1. Firms that proactively use iden-tity-claiming mechanisms (i.e., templates, sto-ries, and leadership signals) are more likely tobecome the cognitive referents in distinctmarkets.
Proposition 2. Firms that proactively use de-marcating alliances with established firms(i.e., revenue sharing, equity investment, anti-leader positioning) are more likely to facelower levels of competition.
Proposition 3. Firms that proactively use con-trolling acquisitions of entrepreneurial rivals(i.e., elimination, market coverage, entry block-ing) are more likely to have higher marketshare.
More broadly, our findings offer a holistic (i.e.,relatively complete and integrated) view of howentrepreneurs shape organizational boundaries.Rather than optimizing atomistic boundary deci-sions of limited importance, successful entrepre-neurs interrelate decisions to form patterns ofboundary processes. These boundary processes arereinforcing because they operate in interrelated do-mains—cognitions in the emerging field, relationsof key players with the nascent market, and re-source distribution in the market. By operating inmultiple domains, entrepreneurial actors broadentheir repertoires of strategic actions to shape themarket and achieve dominance. For example, alli-ances parry potential competitors that are too largeto acquire, while acquisitions eliminate actual
smaller rivals that are affordable. Since both mech-anisms restrict competition and create a favorableindustry structure, they enhance the competitiveposition of the young firm.
Interestingly, combinations of these boundaryprocesses are often synergistic. For example, thesuccessful claiming activities at Secret enhancedthe market profile of the firm and so increased thelikelihood of enticing established firms to form al-liances that demarcated the market. In turn, suc-cessful alliance formation made Secret a more cen-tral actor in the nascent market, reinforcing itsclaiming activities. Acquisitions carried out to con-trol the market were made easier by the claimingactivities that made Secret a more attractive buyer.Its demarcating alliances also eliminated some ri-val buyers for the targeted ventures. Thus, the threeboundary processes were closely intertwined andenabled Secret to achieve a dominant position in adistinct market. In contrast, Magic executives didnot support their successful claiming process witheither demarcating alliances or controlling acquisi-tions. As a result, they faced intense competitivepressures and pricing wars. Only their very largeIPO war chest, which was substantially aided bytheir claiming process, prevented failure.
In summary, this holistic view suggests thatboundary processes depend on and reinforce eachother, operating in different but interrelated do-mains of the institutional fabric (e.g., cognitions,relations, and resources). As a result, they are in-teractive and synergistic, not atomistic.
Proposition 4. Entrepreneurs that intertwineboundary processes are more likely to (a) be-come the cognitive referents in distinct mar-kets, (b) face lower levels of competition, and(c) have higher market share.
Further, our analysis points to the key insightthat power is the unifying boundary logic. Al-though other logics are present in our data, poweris dominant. For example, though claiming activitiescan sometimes enhance legitimacy, which is consis-tent with an identity-based logic (Lounsbury & Glynn,2001), our findings emphasize the power logic ofachieving dominance by becoming the cognitive ref-erent in a market. Similarly, while acquisitions andalliances add resources and are sometimes under-taken with competence and/or efficiency logics inmind, they too are often motivated by a strategicinterest in competitive dominance. An informantconfirmed the role of power in driving decisions:
We felt that we were going to get more marketpower, which was our goal, and pick up the scaleeconomies. But we wouldn’t have done it just for
662 AugustAcademy of Management Journal
scale economies [efficiency logic]. Market powerwas the driving reason.
Our inductive analysis reveals that entrepreneursactually wield power by relying on a strategy ofusing soft power. Recent work in political science(Nye, 2004) distinguishes “hard power,” which isbased on coercion, direct rewards, and extensiveresource deployment to force others’ behaviors,from “soft power,” which is based on subtle influ-ence mechanisms that cause others to willinglybehave in ways that benefit the focal agent. Man-agement scholars have used this distinction to de-scribe effective behavior toward complementers(Yoffie & Kwak, 2006). Similarly, we adopt theconcept of soft power because it captures well thestrategy and underlying tactics that entrepreneursin young ventures use to construct and dominatenascent markets.
One such power tactic that we observed is illu-sion. Illusion is the use of deception, such asshielding intentions and exaggerating one’s impor-tance to gain advantage. For example, emittingleadership signals (for instance, Secret’s active set-ting of industry standards despite having only 20employees and one lawyer) portray a firm’s accom-plishments and resources as greater than they ac-tually are. Alliances are used to pretend to estab-lished firms that ventures are real options and opento a later acquisition when entrepreneurs have nointention of selling. Similarly, in acquisitions anillusion-wielding buying firm emphasizes synergyand common culture while obscuring its predatoryintent.
A second power tactic is exploitation of the ten-dencies of others. Rather than attempting to forceother actors to act in a desired way, entrepreneursexploit their natural tendencies. For example, sto-ries (e.g., the romantic story about Haven’s found-ing) exploit people’s tendency to be overly influ-enced by vivid stories with personal details(Nisbett & Ross, 1980; Rindova, Pollock, & Hay-ward, 2006). Similarly, offering their firms as part-ners to established firms in alliances that demar-cate the market exploits the tendency of establishedfirms to delay entry until market attractiveness isestablished (Markides & Gerosky, 2005). Likewise,the use of acquisitions to blunt the competitiveimpact of young rivals exploits the preference ofentrepreneurs to be bought by other entrepreneurs(Graebner & Eisenhardt, 2004).
A third power tactic is timing (either preemptiveor delaying). Like the executives of establishedfirms (Katila & Chen, 2008), entrepreneurs strategi-cally use asynchronous timing to their advantage.For example, preemptive use of templates can lock
in a focal firm as the cognitive referent in its nas-cent market before other firms take action (e.g.,Magic). Preemptive alliances can tie up would-becompetitors in contractual obligations before theyfully realize the market’s potential (e.g., Secret).Preemptive acquisitions can eliminate stepping-stones into the market before others can use them.Similarly, delaying can be advantageous (for exam-ple, Midway used acquisitions to buy rivals andgain patents that delayed entry by large competi-tors; Haven delayed negotiations with potentialbuyers to gain position).
Taken together, these soft-power tactics offer in-sight into why the boundary processes that we ob-serve are effective, and how they enable new firmsin nascent markets to gain advantage and evendominance over established firms and entrepre-neurial rivals:
Proposition 5. Firms that use soft-power tacticsto shape boundaries (i.e., illusion, exploitingothers’ natural tendencies, timing) are morelikely to achieve (a) cognitive dominance (be-come the cognitive referent in a distinct mar-ket) and (b) competitive dominance (face alower level of competition, have greater marketshare).
Finally, our emergent theoretical model suggeststhat these boundary processes enable firms to con-struct distinct markets in which they sustain nearmonopolies. By combining processes from differentdomains of action (i.e., cognitions, relations, re-sources) and intertwining market and firm bound-aries, entrepreneurs can structure a market space totheir advantage and develop a dominant, sustain-able position. For example, Secret’s proactive useover time of claiming, demarcating, and controllingprocesses contributed substantially to its ability todevelop and sustain a near-monopoly position inits constructed markets, becoming the cognitive ref-erent, with very few competitors and an over 80percent market share after five years. In contrast,Saturn’s less consistent and effective use of thesesame processes hampered its ability to establish adistinct market and remain its clear leader.
Proposition 6. Firms that, over time, proac-tively combine claiming, demarcating, andcontrolling boundary processes are more likelyto sustain near-monopoly positions in con-structed markets.
A key issue is whether post hoc “sensemaking”influences our findings. As noted earlier, we re-duced this possibility by triangulating real-time in-formation with retrospective accounts and by usinga data collection approach that reduces informant
2009 663Santos and Eisenhardt
bias (Huber, 1985; Huber & Power, 1985). In fact,only a few informants could describe the “big pic-ture” of what had occurred at their firms. Thus, itseems unlikely that informants with different infor-mation, focused on different decisions at variedtimes, and at very different firms with diverse start-ing positions, would have exhibited similar retro-spective sensemaking. Finally, our findings do notrequire hyperrationality among the entrepreneurs.Although they shared the central challenge of suc-ceeding in nascent markets, our entrepreneurs of-ten approached this challenge by taking actions asevents unfolded and learning from mistakes. Thus,our entrepreneurs plausibly described a blend ofemergent and deliberate actions, together with mis-takes and serendipitous learning that occurred whilethey were trying to succeed in nascent markets.
DISCUSSION
We add to the study of organizational boundariesand market construction. Our core theoretical con-tribution is a holistic framework of the longitudinalprocesses by which successful entrepreneurs shapeorganizational boundaries and construct new mar-kets. Table 5 summarizes this framework. Theseprocesses rely on boundary mechanisms operatingin three distinct domains: cognitions (i.e., shapingmeanings and becoming the cognitive referent in amarket), relations (i.e., bounding and defending themarket by creating industry roles for powerful oth-ers), and resources (i.e., owning as much of themarket space as possible). Collectively, they expli-cate how entrepreneurs interrelate decisions overtime to shape boundaries and construct markets totheir advantage.
A more fundamental contribution is the reinvig-oration of interorganizational power. We find thatpower is the underlying theoretical logic—that is,successful entrepreneurs adopt a monopolistic im-perative of dominating a market that they construct.They approach this imperative by linking their organ-izational boundaries with market construction usingsoft-power strategies. As Alvarez and Barney (2007)argued, a central debate in entrepreneurship iswhether new markets constitute opportunities wait-ing to be discovered by alert entrepreneurs (Kirz-ner, 1997; Shane & Venkataraman, 2000) orwhether new markets are actually willed into exis-tence by savvy entrepreneurs (Rindova & Fombrun,2001; Sarasvathy, 2001). Although opportunitiesmatter, we find that entrepreneurs engaging inpower-based, firm-centered strategies can play acritical role in shaping new markets.
Ties to Institutional Entrepreneurship
Responding to calls for studying interest andagency (DiMaggio, 1988; Fligstein, 1997), scholarshave extended new institutional theory from con-formity to entrepreneurial action. In particular, in-stitutional entrepreneurs are those individuals andorganizations that engage in creative activities toproduce or change existing institutional arrange-ments (Hwang & Powell, 2005). Thus far, institu-tional entrepreneurship research has focused oncollective actions by which industry groups (Rao,2004), state actors and nongovernmental organiza-tions (Maguire & Hardy, 2006), corporate elites(Greenwood & Suddaby, 2006), and social move-ments (Sine & David, 2003) legitimate new organi-zational forms and institutions, often using the me-dia (Kennedy, 2005; Rosa, Porac, Runser-Spanjol, &Saxon, 1999). Our contribution is to emphasize howindividual entrepreneurial firms (not collectives) thatare peripheral (not central or elite) in existing insti-tutional fields use a broad range of boundary mecha-nisms to construct new markets and become domi-nant (not just legitimate) firms.
Our model is consistent with elements identifiedin institutional research, namely, the importance of(1) analogies for “sensegiving” in nascent markets(Gavetti et al., 2005; Leblebici et al., 1991), (2) abalance between familiarity and novelty (Hargadon& Douglas, 2001; Maguire & Hardy, 2005), and (3)the role of stories (Lounsbury & Glynn, 2001). Likethe models of others (Suddaby & Greenwood,2005), our model emphasizes language and rhetor-ical framing. However, while institutional theoryfocuses on legitimacy and cognitive-based strate-gies, we focus on dominance and integrate cogni-tive and competitive strategies. In particular, legit-imacy centers on winning a contest to become theappropriate institutional field or organizationalform (Suchman, 1995); dominance centers on howan individual firm wins both cognitive and eco-nomic competitions. Our contribution is thus a ho-listic view that infuses new institutional theorywith themes of power, self-interest, and dominanceat the firm level, themes that ironically began in theoften neglected “old” institutional theory (Selz-nick, 1949). We show how seemingly disparateboundary strategies are interrelated, why they areeffective, and how actors use them strategically tobecome dominant firms, not just legitimate forms.
The Role of Ambiguity in Market Creation
The entrepreneurial actions that we identify arefacilitated by the ambiguity of nascent markets. Thefluid social structure and multiple possible mean-
664 AugustAcademy of Management Journal
ings of events in nascent markets make audiencesreceptive to new interpretations that reduce ambi-guity. Interestingly, Weick argued that two reac-tions to ambiguity are possible: “Ambiguity under-stood as confusion created by multiple meaningscalls for social construction and invention. Ambi-guity understood as ignorance created by insuffi-cient information calls for more careful scanningand discovery” (1995: 95). Building on this insight,we suggest that it is in the nature of entrepreneurs
to deal with ambiguity through social construction,and it is in the nature of executives in establishedfirms to react to ambiguity by scanning for moreinformation. Thus, the latter may be open to andeven seek the “sensegiving social construction”proposed by entrepreneurs. This notion leads to anintriguing view of nascent markets as competitivefields for alternative conceptions espoused byyoung firms vying for dominance, even as estab-lished firms wait on the sidelines for the outcome.
TABLE 5Framework for Constructing Markets
Scope conditions: High-ambiguity environment (e.g., nascent or destabilized market)● Audiences are trying to make sense of the market situation and may be open to the structuring actions of entrepreneurs who try to
enact a particular market conception.● Nonexistent or fluid structure facilitates the moves of entrepreneurs to shape cognitions, relations and resources in the market to
their advantage.
Elements Claiming a Market Demarcating the Market Controlling the Market
Domain of actions Cognitions Relationships ResourcesObjective Become cognitive referent
in a distinct market.Determine market perimeter,
and define industry structureand roles for powerfulplayers.
Cover market space, andeliminate or delay rivalry.
Dominant logic Sensegiving Co-optation OwnershipOrganizational capability Shaping and promoting
identityDeveloping alliances Making acquisitions
Mechanisms Templates Equity investments Eliminate competing modelsLeadership signals Revenue-sharing agreements Increased market coverageStories Antileader positioning Block entry of established
firmsQuotes illustrating
entrepreneurial mind-set in nascent markets
“At the time, it was the wildwest—there was no playbook for the Internet orour space—we created theplaybook.”
“Because we knew that bythe end of 2000, we wouldpretty much have definedwhat the company stoodfor in customers’ minds.. . . You have to do thatstuff pretty fast otherwiseby the time you getaround to do it . . . youcan’t change people’sperceptions about whatyou are.”
“We had a combination of afortunate timing equationand a focused objective inwhat becomes—whenhistory is written—afundamentally differentmarket.”
“One of the things that separateus is that we are alwaysworried about who could takethis away from us and wetend to find a way tocooperate when there is awin-win scenario. Before theyrecognize us and turn theirattention to us, we find a wayfor them to benefit from theirassociation with us.”
“So we have kind of alsocreated a demilitarized zonefor ourselves. We won’t goabove this area and get intheir face on their mainstreamproduct line. I think that wasvery important.”
“We were not afraid of ourcompetitors. We were alwaysafraid of the biggerpeople. . . . We tried to getthem as customers becausewe didn’t want them to say:“Hey, this is a lucrativemarket, let’s get in there andcompete against them.”
“We felt that we were going toget more market power,which was our goal, andpick up the scaleeconomies. We would nothave done it just for scaleeconomies. Market powerwas the driving reason.”
“The primary reason was tomake sure that neither ‘A’or ‘B’ companies ended upwith the technology. . . .Their products today arebetter than the old stuff butthey can’t go all the way(without this technology).So it was a blocking moveon our part.”
“The real concern was theprospect that a largercompany might buy thisventure and use it to launcha competing service. Theseconcerns were well founded.Harbor did not know it, buta large player was makingsuch an overture to theventure at the time.”
2009 665Santos and Eisenhardt
Yet establishing a winning market conception isa broader endeavor than, for example, establishinga new product concept (Garud & Rappa, 1994). Amarket conception includes structuring industryroles around an ambiguous market through negoti-ation of boundaries with powerful incumbents inproximate markets. This “structuration” differsfrom that suggested by others. For example, White(2002) argued that the key to understanding eco-nomic action is that producers seek market nichesto maximize profit and minimize competition, aview that is consistent with our model. However,White argued that firms construct these markets byfinding roles in relation to rivals to create a peckingorder of quality (White, 1981, 1992). Instead, wefind that entrepreneurs largely ignore their closerivals and proactively focus on establishing a struc-ture of roles (e.g., suppliers, customers, investors,complementers) for powerful but more distantfirms. In this regard, entrepreneurs become activeshapers of a new industry architecture by definingboundaries and division of labor (Jacobides, Knud-sen, & Augier, 2006; Ozcan & Eisenhardt, 2009).These actions, of course, require sharing informa-tion and regular resource exchanges, and can leadto dependence. Still, the key insight is that, despiteoften weak initial resources, entrepreneurs havethe ability to engage established players, constructa market with a favorable industry structure, andachieve dominance. This ability is surprising andrests on the clever use of power.
Power as Dominant Logic
A key contribution of our work is a reinvigoratedview of interorganizational power. Unexpectedly,we identify power as the dominant logic for bound-ary formation in nascent markets. The theoreticalfoundations of interorganizational power stemfrom classical work in organizational sociology andindustrial organization. Work such as Selznick’s(1949) study of co-optation, Thompson’s insightson interdependence (1967), Zald’s (1970) focus onopen systems, and Pfeffer and Salancik’s (1978)resource dependence theory emphasizes how or-ganizations attempt to control key exchange rela-tions to reduce the uncertainty they face. Work inindustrial organization (IO) economics, such asBain’s (1956) analysis of barriers to competitionand entry deterrents, and Porter’s conversion ofthese ideas into a manual for strategic action (Caves& Porter, 1977; Porter, 1980), emphasizes how themanipulation of market structure by powerful firmscan create sustainable advantage. Despite thestrength of these ideas, as Mizruchi and Yoo notedin their review (2002: 602, 614), research on interor-
ganizational power has stalled. They argued thatcontemporary research rests on distant, archivaldata that offer little insight into how firms actuallyexercise power. An in-depth field study such asours thus provides an opportunity to reinvigoratepower. Interestingly, we found that our private in-terview data, which were consistent with the ac-tions taken by the firms, often revealed an anticom-petitive power logic that was missing from archivaldata, like media reports and press releases, whereboundary decisions were benignly framed usingvalue creation arguments. This bias in archival datamay help explain the novelty of our insights regard-ing the monopolistic imperative of entrepreneurs.
We contribute to a reinvigorated view of powerin three ways. First, we extend theories of power toambiguous environments. Resource dependenceand IO economics focus on managing uncertaintyin structured environments—that is, reducing thedependence of a focal firm on external forces. Incontrast, we focus on ambiguous environmentswhere there is little or no market structure, no clearmeaning, and unknown dependence. Here, entre-preneurs attempt to reduce ambiguity while shap-ing a favorable market structure before others do it(Davis et al., in press). But to succeed, entrepre-neurs must engage powerful others, and so oftenincrease dependence. Of course, the traditionalstrategy of reducing dependence (Pfeffer & Salan-cik, 1978) would just isolate a venture and probablydoom it. Thus, entrepreneurs need first to tradeambiguity (i.e., lack of meaning and structure) foruncertainty (i.e., dependence on others in a knownstructure that they help shape). Addressing the re-sulting uncertainty may then involve using tradi-tional resource dependence mechanisms (Casciaro& Piskorski, 2005; Gulati & Sytch, 2007; Katila etal., 2008). In summary, we extend resource depen-dence to ambiguous settings by noting (1) an initialstrategy of reducing ambiguity through favorablemarket structuring and (2) a later strategy, enactedafter the market crystallizes, of managing the result-ing uncertainty as suggested by current theory.
Second, we extend current theories of power toentrepreneurial firms. A key conundrum in IO eco-nomics and resource dependence theories is thatthey are one-sided and so do not fully resolve theissue of why one firm would accept reducing thedependence of another, particularly when depen-dence is asymmetric (Casciaro & Piskorski, 2005).Theorists have usually advocated hard-power tac-tics feasible for resource-rich firms in establishedmarkets, such as making major resource commit-ments to create entry barriers, threatening preda-tory pricing strategies that cut profits for all, andinterlocking boards with powerful peers (Caves &
666 AugustAcademy of Management Journal
Porter, 1977; Pfeffer & Salancik, 1978; Porter, 1980).These tactics are unrealistic for entrepreneurialfirms. Rather, ventures usually begin with few re-sources, suffer from a “liability of newness,” andhave a “strategic manual” that advises conformityto established recipes and legitimating ties withhigh-status firms. Our contribution to resource de-pendence and other theories of power is to high-light the soft-power tactics of entrepreneurs—illu-sion, timing, and exploitation of others’ naturaltendencies—and to explain their effectiveness inshaping and dominating nascent markets.
Third, we note that well-known mechanisms (i.e.,identity, alliances, and acquisitions) that are notoften associated with power are actually part of afirm’s strategic arsenal for creating dominance. Forexample, although identity can be a sense-makingdevice for organization members (Corley & Gioia,2004; Dutton & Dukerich, 1991), it can also be asense-giving device to help a firm win the contestto be the cognitive referent in a market. Similarly,although alliances can access resources (Eisenhardt& Schoonhoven, 1996; Gulati, 1998), they can also
co-opt potential rivals and favorably shape anemerging industry structure. Although acquisitionscan bring in valuable resources (Ahuja & Katila,2001; Capron et al., 1998), they can also block oth-ers’ access to resources, destroy threatening re-sources, and eliminate competition. This latterfinding brings a fresh perspective on acquisitionperformance. Acquisitions are often seen as failureswhen they do not lower costs or yield new busi-nesses. Yet this reasoning fails to include what isclear using a power lens: acquisitions can be verysuccessful when they help firms to dominate theirmarkets. We thus offer a reminder that these com-mon mechanisms (i.e., identity, alliance, and ac-quisition) often have power as a key strategic ra-tionale. Table 6 outlines our framework for the useof power by entrepreneurs in nascent markets.
Overall, we contribute to a reinvigorated view ofinterorganizational power by offering a strategic the-ory of power that blends institutional (cognitive) andresource dependence (competitive) lenses. Ourmodel is based on specific strategic actions that reston the nuances of soft power. It is thus uniquely
TABLE 6Framework for the Use of Power by Entrepreneurs in Nascent Markets
Lacking traditional “hard” sources of power such as large scale, deep pockets, and a strong customer base, entrepreneurial firms caninfluence other market actors using “soft” sources of power that reduce market ambiguity and influence the actions of others.
Claiming a Market Demarcating the Market Controlling the Market
Power logic Identity-based actions can beused as sensegivingdevices.
Alliances can be used to co-optpowerful players, not justgain resources.
Acquisitions can be used tocancel smaller rivals andblock stepping-stones forlarger ones.
ambiguity for others.Identify potential competitors
early to preempt their entry.Anticipate ventures with most
threatening competingmodels.
Illusion Provide symbols oflegitimacy and power.
Provide illusion of being a“real option” for partners.
Disguise intentions to shutteracquired targets.
Exploit tendencies of others Add familiarity into a novelproduct or service as userstrust what is familiar.
Give sufficient incentive tokeep partners from strayinggiven their preference tofocus on their own coremarkets.
Take advantage of venturespreference forentrepreneurial buyers, andestablished playerspreference for market entrythrough acquisitions.
Potential implementationpitfalls
Too late in creating identity.Identity is unfamiliar or
boring.Identity is too close to
nearby established market.
Underestimate threat ofestablished firms.
Provide too little incentive topartners to stay loyal.
Choose wrong partners, e.g.,small, young ventures.
Too few financial resources tobuy most threatening targets.
Too late to make blockingacquisitions.
Targets refuse to sell.
2009 667Santos and Eisenhardt
suited to understanding entrepreneurial actors withhigh aspirations in ambiguous settings and may ex-plain how young firms are able to dominate nascentmarkets. This repertoire of actions may, in fact, con-stitute their primary strategy.
Conclusion
Our aim was to gain a holistic, longitudinal un-derstanding of boundary work. We thus studiedfirms that were longer-lived, more visible, andmore successful than many ventures. With regardto generalizability, it is critical to note that thesefirms operated in distinct areas and had very dif-ferent starting conditions. In addition, not all firmsbegan with superior resources, and most made ma-jor mistakes that added useful variance to ourstudy. More significantly, the processes and under-lying soft-power strategies that we observed arerealistic for many entrepreneurs, albeit at a morelocal scale (such as claiming a market in a metro-politan area or small niche, allying with otherneighborhood businesses, or consolidating withlike-minded small firms). Thus, although we stud-ied unusual firms, the strategic approach that weunveil may generalize to other ventures with highaspirations in nascent markets. The next step isempirical testing.
To conclude, our most important contribution isbringing power back into the discourse on strategyand organizations, particularly for entrepreneurialfirms in nascent markets. Although recent researchhas emphasized social networks, dynamic capabil-ities, legitimacy, and innovation, we offer a strongreminder that agency and strategic action often reston the rationale of power.
REFERENCES
Ahuja, G., & Katila, R. 2001. Technological acquisitionsand the innovation performance of acquiring firms.Strategic Management Journal, 15: 103–120.
Aldrich, H. E., & Fiol, C. M. 1994. Fools rush in? Theinstitutional context of industry creation. Academyof Management Review, 19: 645–670.
Aldrich, H. E., & Ruef, M. 2006. Organizations evolving(2nd ed.). London: Sage.
Alvarez, S. A., & Barney, J. B. 2007. Discovery and cre-ation: Alternative theories of entrepreneurial action.Strategic Entrepreneurship Journal, 1(1): 11–26.
Argyres, N. 1996. Evidence on the role of firm capabili-ties in vertical integration decisions. Strategic Man-agement Journal, 17: 129–150.
Bain, J. S. 1956. Barriers to new competition. Cam-bridge, MA: Harvard University Press.
Bingham, C., Eisenhardt, K., & Davis, J. 2009. Openingthe black box: What firms explicitly learn fromtheir process experience. Working paper, Univer-sity of North Carolina.
Brusoni, S., Prencipe, A., & Pavitt, K. 2001. Knowledgespecialization, organizational coupling, and theboundaries of the firm: Why do firms know morethan they make? Administrative Science Quarterly,46: 597–621.
Burton, M. D., & Beckman, C. M. 2007. Leaving a legacy:Role imprints and succession turn-over in youngfirms. American Sociological Review, 72: 239–266.
Capron, L., Dussauge, P., & Mitchell, W. 1998. Resourceredeployment following horizontal acquisitions inEurope and North America, 1988–1992. StrategicManagement Journal, 19: 631–661.
Casciaro, T., & Piskorski, M. J. 2005. Power imbalance,mutual dependence, and constraint absorption: Acloser look at resource dependence theory. Admin-istrative Science Quarterly, 50: 167–199.
Caves, R. E., & Porter, M. E. 1977. From entry barriers tomobility barriers: Conjectural decisions and con-trived deterrence to new competition. QuarterlyJournal of Economics, 91: 241–261.
Chatain, O., & Capron, L. 2008. Competitors’ resource-oriented strategies: Acting upon competitors’ re-sources through interventions in factor markets andpolitical markets. Academy of Management Re-view, 33: 97–121.
Corley, K. G., & Gioia, D. A. 2004. Identity ambiguity andchange in the wake of a corporate spin-off. Admin-istrative Science Quarterly, 49: 173–208.
David, R. J., & Han, S.-K. 2004. A systematic assessmentof the empirical support for transaction cost econom-ics. Strategic Management Journal, 25: 39–58.
Davis, M. S. 1971. That’s interesting: Towards a phenom-enology of sociology and a sociology of phenomenol-ogy. Philosophy of Social Sciences, 1: 309–344.
Davis, J., Eisenhardt, K. M., & Bingham, C. B. In press.Optimal structure, market dynamism and the strat-egy of simple rules. Administrative Science Quar-terly.
DiMaggio, P. J. 1988. Interest and agency in institutionaltheory. In L. G. Zucker (Ed.), Institutional patternsand organizations: Culture and environment:3–21. Cambridge, MA: Ballinger.
Dougherty, D. 1990. Understanding new markets fornew products. Strategic Management Journal,11: 59 –78.
Dutton, J. E., & Dukerich, J. M. 1991. Keeping an eye onthe mirror: Image and identity in organizational ad-aptation. Academy of Management Journal, 34:517–554.
Dyer, J. H. 1996. Does governance matter: Keiretsu alli-ances and asset specificity as sources of Japanese
Dyer, J. H., & Singh, H. 1998. The relational view: Coop-erative strategy and sources of interorganizationalcompetitive advantage. Academy of ManagementReview, 23: 660–679.
Eisenhardt, K. 1989a. Making fast strategic decisions inhigh-velocity environments. Academy of Manage-ment Journal, 32: 543–576.
Eisenhardt, K. M. 1989b. Building theories from casestudy research. Academy of Management Review,14: 532–550.
Eisenhardt, K. M., & Graebner, M. E. 2007. Theory build-ing from cases: Opportunities and challenges. Acad-emy of Management Journal, 50: 25–32.
Eisenhardt, K. M., & Schoonhoven, C. B. 1996. Resource-based view of strategic alliance formation: Strategicand social effects in entrepreneurial firms. Organi-zation Science, 7: 136–150.
Fiol, C. M. 1995. Corporate communications: Comparingexecutives’ private and public statements. Academyof Management Journal, 38: 522–536.
Fligstein, N. 1997. Social skill and institutional theory.American Behavioral Scientist, 40: 397–405.
Garud, R., & Rappa, M. A. 1994. A socio-cognitive modelof technology evolution: The case of cochlear im-plants. Organization Science, 5: 344–362.
Gavetti, G., Levinthal, D. A., & Rivkin, J. W. 2005. Strat-egy making in novel and complex worlds: The powerof analogy. Strategic Management Journal, 26:691–712.
Gilbert, C. 2005. Unbundling the structure of inertia:Resource versus routine rigidity. Academy of Man-agement Journal, 48: 741–763.
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.
Graebner, M. E. 2004. Momentum and serendipity: Howacquired leaders create value in the integration oftechnology firms. Strategic Management Journal,25: 751–777.
Graebner, M. 2009. Caveat venditor: Trust asymmetriesin acquisitions of entrepreneurial firms. Academy ofManagement Journal, 52: 435–472.
Graebner, M. E., & Eisenhardt, K. M. 2004. The seller’sside of the story: Acquisition as courtship and gov-ernance as syndicate in entrepreneurial firms. Ad-ministrative Science Quarterly, 49: 366–403.
Greenwood, R., & Suddaby, R. 2006. Institutional entre-preneurship in mature fields: The big five account-ing firms. Academy of Management Journal, 49:27–48.
Gulati, R. 1998. Alliances and networks. Strategic Man-agement Journal, 19: 293–317.
Gulati, R., & Higgins, M. C. 2003. Which ties matterwhen? The contingent effects of interorganizationalpartnerships on IPO success. Strategic Manage-ment Journal, 24: 127–144.
Gulati, R., & Sytch, M. 2007. Dependence asymmetry andjoint dependence in interorganizational relation-ships: Effects of embeddedness on a manufacturer’sperformance in procurement relationships. Admin-istrative Science Quarterly, 52: 32–69.
Hallen, B. L. 2008. The causes and consequences of theinitial network positions of new organizations: Fromwhom do entrepreneurs receive investments? Ad-ministrative Science Quarterly, 53: 685–718.
Hargadon, A. B., & Douglas, Y. 2001. When innovationmeets institutions: Edison and the design of the elec-tric light. Administrative Science Quarterly, 46:476–501.
Harris, S., & Sutton, R. I. 1986. Functions of partingceremonies in dying organizations. Academy ofManagement Journal, 29: 5–30.
Heath, C., Bell, C., & Sternberg, E. 2001. Emotional selec-tion in memes: The case of urban legends. Journal ofPersonality and Social Psychology, 81: 1028–1041.
Hsu, G., & Hannan, M. T. 2005. Identities, genres, andorganizational forms. Organization Science, 16:474–490.
Huber, G. P. 1985. Participant descriptions of organiza-tional decisions: Temporal stability and responseorder biases. Academy of Management Journal, 28:943–950.
Huber, G. P., & Power, D. J. 1985. Retrospective reports ofstrategic-level managers: Guidelines for increasingtheir accuracy. Strategic Management Journal, 6:171–180.
Hwang, H., & Powell, W. 2005. Institutions and Entrepre-neurship. In S. A. Alvarez, R. Agarwal, & O. Soren-son (Eds.), Handbook of entrepreneurship re-search: 179–210. New York: Springer.
Jacobides, M., & Billinger, S. 2006. Designing the bound-aries of the firm: From “make, buy or ally” to thedynamic benefits of vertical architecture. Organiza-tion Science, 17: 249–261.
Jacobides, M. G., & Hitt, L. M. 2005. Losing sight of theforest for the trees? Productive capability and gainsfrom trade as drivers of vertical scope. StrategicManagement Journal, 26: 1209–1227.
Jacobides, M. G., Knudsen, T., & Augier, M. 2006. Bene-fiting from innovation: Value creation, value appro-priation and the role of industry architectures.Research Policy, 35: 1200–1221.
Kaplan, S., & Murray, F. Forthcoming. Entrepreneurshipand the construction of value in biotechnology. InResearch on the sociology of organizations.
Kaplan, S., & Tripsas, M. 2008. Thinking about technol-ogy: Applying a cognitive lens to technical change.Research Policy, 37: 790–805.
2009 669Santos and Eisenhardt
Katila, R., & Chen, E. 2008. Effects of search timing oninnovation: The value of not being in sync withrivals. Administrative Science Quarterly, 53: 593–625.
Katila, R., Rosenberger, J., & Eisenhardt, K. 2008. Swim-ming with sharks: Technology ventures defensemechanisms and corporate relationships. Adminis-trative Science Quarterly, 53: 295–332.
Kennedy, M. C. 2005. Behind the one-way mirror: Refrac-tion in the construction of product market catego-ries. Poetics, 33: 201–226.
Kirzner, I. M. 1997. Entrepreneurial discovery and thecompetitive market process: An Austrian approach.Journal of Economic Literature, 35: 60–85.
Kumar, N., Stern, L. W., & Anderson, J. C. 1993. Conduct-ing interorganizational research using key infor-mants. Academy of Management Journal, 36:1633–1651.
Larsson, R., & Finkelstein, S. 1999. Integrating strategic,organizational, and human resource perspectives onmergers and acquisitions: A case survey of synergyrealization. Organization Science, 10: 1–26.
Leblebici, H., Salancik, G. S., Copay, A., & King, T. 1991.Institutional change and the transformation of interor-ganizational fields: An organizational history of theU.S. radio broadcasting industry. AdministrativeScience Quarterly, 36: 333–363.
Lounsbury, M., & Glynn, M. A. 2001. Cultural entrepre-neurship: Stories, legitimacy, and the acquisition ofresources. Strategic Management Journal, 22: 545–564.
Maguire, S., & Hardy, C. 2005. Identity and collaborativestrategy in the Canadian HIV/AIDS treatment do-main. Strategic Organization, 3: 11–45.
Maguire, S., & Hardy, C. 2006. The emergence of newglobal institutions: A discursive perspective. Orga-nization Studies, 27: 7–29.
Markides, C., & Gerosky, P. A. 2005. Fast second: Howsmart companies bypass radical innovation to en-ter and dominate new markets. San Francisco: Jos-sey-Bass.
Miles, M. B., & Huberman, A. M. 1994. Qualitative dataanalysis: An expanded sourcebook (2nd ed.).Thousand Oaks, CA: Sage.
Miller, C. C., Cardinal, L. B., & Glick, W. H. 1997. Retro-spective reports in organizational research: A reex-amination of recent evidence. Academy of Manage-ment Journal, 40: 189–204.
Mizruchi, M. S., & Yoo, M. 2002. Interorganizationalpower and dependence. In J. Baum (Ed.), Compan-ion to organizations: 599–620. Oxford, U.K.: Black-well.
Nisbett, R., & Ross, L. 1980. Human inference: Strate-gies and shortcoming of social judgment. Engle-wood Cliffs, NJ: Prentice Hall.
Nye, J. S. 2004. Soft power: The means to success inworld politics. Cambridge, MA: Public Affairs.
Ozcan, C. P., & Eisenhardt, K. M. 2009. Origins of allianceportfolios: Entrepreneurs, network strategies, andfirm performance. Academy of Management Jour-nal, 52: 246–279.
Penrose, E. 1995. The theory of the growth of the firm(3rd ed.). Oxford, U.K.: Basil Blackwell.
Peteraf, M. A. 1993. The cornerstones of competitiveadvantage: A resource-based view. Strategic Man-agement Journal, 14: 179–191.
Pfeffer, J., & Salancik, G. 1978. The external control oforganizations: A resource dependence perspec-tive. New York: Harper & Row.
Poppo, L., & Zenger, T. 1998. Testing alternative theoriesof the firm: Transaction cost, knowledge-based, andmeasurement explanations for make-or-buy deci-sions in information services. Strategic Manage-ment Journal, 19: 853–877.
Porac, J. F., Thomas, H., Wilson, F., Paton, D., & Kanfer,A. 1995. Rivalry and the industry model of Scottishknitwear producers. Administrative Science Quar-terly, 40: 203–227.
Porac, J. F., Ventresca, M. J., & Mishina, Y. 2002. Interor-ganizational cognition and interpretation. In J. C.Baum (Ed.), Companion to organizations: 579–598:London: Blackwell.
Porter, M. E. 1980. Competitive strategy. New York: FreePress.
Powell, W. W., Koput, K. W., & Smith-Doerr, L. 1996.Interorganizational collaboration and the locus ofinnovation: Networks of learning in biotechnology.Administrative Science Quarterly, 41: 116–145.
Rao, H. 1994. The social construction of reputation: Cer-tification contests, legitimation, and the survival oforganizations in the American automobile industry:1895–1912. Strategic Management Journal, 15: 29–44.
Rao, H. 2004. Institutional activism in the early Ameri-can automobile industry. Journal of Business Ven-turing, 19: 359–384.
Rindova, V., & Kotha, S. 2001. Continuous “morphing”:Competing through dynamic capabilities, form, andfunction. Academy of Management Journal, 44:1263–1280.
Rindova, V., Pollock, T., & Hayward, M. L. A. 2006.Celebrity firms: The social construction of marketpopularity. Academy of Management Review, 31:50–71.
Rindova, V. P., & Fombrun, C. J. 2001. Entrepreneurialaction in the creation of the specialty coffee niche. InC. B. Schoonhoven & E. Romanelli (Eds.), The en-trepreneurship dynamic: 236–261. Stanford, CA:Stanford University Press.
Rosa, J. A., Porac, J. F., Runser-Spanjol, J., & Saxon, M. S.
670 AugustAcademy of Management Journal
1999. Sociocognitive dynamics in a product market.Journal of Marketing, 63(4): 64–77.
Santos, F. M., & Eisenhardt, K. M. 2005. Organizationalboundaries and theories of organization. Organiza-tion Science, 16: 491–508.
Sarasvathy, S. 2001. Causation and effectuation: Towarda theoretical shift from economic inevitability to en-trepreneurial contingency. Academy of Manage-ment Review, 26: 243–288.
Schwenk, C. R. 1985. The use of participant recollectionin the modeling of organizational decision process.Academy of Management Review, 10: 496–503.
Scott, W. R. 2003. Organizations: Rational, natural andopen systems (5th ed.). Upper Saddle River, NJ:Prentice Hall.
Selznick, P. 1949. TVA and the grass roots. Berkeley:University of California Press.
Shane, S., & Venkataraman, S. 2000. The promise ofentrepreneurship as a field of research. Academy ofManagement Review, 25: 217–226.
Siggelkow, N. 2007. Persuasion with case studies. Acad-emy of Management Journal, 50: 20–24.
Sine, W. D., & David, R. J. 2003. Environmental jolts,institutional change, and the creation of entrepre-neurial opportunity in the US electric power indus-try. Research Policy, 32: 185–207.
Sine, W. D., Mitsuhashi, H., & Kirsch, D. A. 2006. Revis-iting Burns and Stalker: Formal structure and newventure performance in emerging economic sectors.Academy of Management Journal, 49: 121–132.
Suchman, M. C. 1995. Managing legitimacy: Strategicand institutional approaches. Academy of Manage-ment Review, 20: 571–610.
Suddaby, R., & Greenwood, R. 2005. Rhetorical strategiesof legitimacy. Administrative Science Quarterly,50: 35–67.
Thompson, J. D. 1967. Organizations in action. NewYork: McGraw Hill.
Tripsas, M. 2009. Technology, identity and inertiathrough the lens of “The Digital Photography Com-pany.” Organization Science, 20: 440–461.
Tripsas, M., & Gavetti, G. 2000. Capabilities, cognitionand inertia: Evidence from digital imaging. StrategicManagement Journal, 21: 1147–1161.
Weick, K. 1995. Sensemaking in organizations. London:Sage.
White, H. C. 1981. Where do markets come from? Amer-ican Journal of Sociology, 87: 517.
White, H. C. 1992. Identity and control. Princeton, NJ:Princeton University Press.
White, H. C. 2002. Markets from networks: Socioeco-nomic models of production. Princeton, NJ, & Ox-ford, U.K.: Princeton University Press.
Williamson, O. E. 1981. The economics of organizations:The transaction cost approach. American Journal ofSociology, 87: 548–577.
Williamson, O. E. 1985. The economic institutions ofcapitalism. New York: Free Press.
Williamson, O. E. 1991. Comparative economic organi-zation: The analysis of discrete structural alterna-tives. Administrative Science Quarterly, 36: 269–296.
Yin, R. K. 1994. Case study research—Design andmethods (2nd ed.). Thousand Oaks, CA: Sage.
Yoffie, D. B., & Kwak, M. 2006. With friends like these:The art of managing complementers. Harvard Busi-ness Review, 36(September): 89–98.
Zald, M. N. 1970. Political economy: A framework forcomparative analysis. In M. N. Zald (Ed.), Power inorganizations: 891–903. Nashville: Vanderbilt Uni-versity Press.
Zott, C., & Huy, Q. N. 2007. How entrepreneurs usesymbolic management to acquire resources. Admin-istrative Science Quarterly, 52: 70–105.
Filipe M. Santos ([email protected]) is an assis-tant professor of entrepreneurship at INSEAD and theacademic director of the INSEAD Social Entrepreneur-ship Initiative. He received his doctorate in organizationsand entrepreneurship from the Management Science &Engineering Department of Stanford University. His re-search interests are in the area of new venture creationand growth, entrepreneurship in nascent markets, andsocial entrepreneurship.
Kathleen M. Eisenhardt ([email protected]) is the S. W.Ascherman M.D. Professor and the codirector of the Stan-ford Technology Ventures Program at Stanford Univer-sity. She received her doctorate from Stanford Univer-sity. Professor Eisenhardt conducts research at the nexusof strategy and organization theory with particular em-phasis on entrepreneurial firms and high-velocity envi-ronments.
