THE PARADOX OF STRETCH GOALS: ORGANIZATIONS IN …performance outcomes as gaps between aspira-tion and current performance elicit and guide effort and persistence (Cyert & March, 1963)

THE PARADOX OF STRETCH GOALS:ORGANIZATIONS IN PURSUIT OF THE

SEEMINGLY IMPOSSIBLE

SIM B. SITKINDuke University

KELLY E. SEENew York University

C. CHET MILLERUniversity of Houston

MICHAEL W. LAWLESSUniversity of Maryland

ANDREW M. CARTONThe Pennsylvania State University

We investigate the organizational pursuit of seemingly impossible goals—commonlyknown as stretch goals. Building from our analysis of the mechanisms through whichstretch goals could influence organizational learning and performance, we offer acontingency framework evaluating which organizations are positioned to benefit fromsuch extreme goals and which are most likely to pursue them. We conclude thatstretch goals are, paradoxically, most seductive for organizations that can least affordthe risks associated with them.

As highlighted by a number of organizationaltheorists, organizations must balance short-termperformance concerns with long-term learning ob-jectives (e.g., Levinthal & March, 1993; Sutcliffe,Sitkin, & Browning, 2000). An organization can en-sure continued survival only by performing well inthe near term while positioning itself for strongperformance in an uncertain future. Although lan-guage and concepts from a number of theoreticaldomains can be used to characterize the balanc-ing of attention between the short term and thelong term, the organizational learning literatureprovides the most directly relevant conceptualiza-tion. From this perspective, organizations must in-vest in activities that “exploit” their known currentcapabilities (i.e., refinement, implementation, and

execution) while also investing in activities that“explore” new, unknown possibilities (i.e., experi-mentation, innovation, playfulness; e.g., Kang,Morris, & Snell, 2007; Levinthal & March, 1993;March, 1991; McGrath, 2001; Sitkin, Sutcliffe, &Schroeder, 1994).

Although exploration is critical for long-termlearning, change, and survival, organizations of-ten have difficulty searching outside their currentroutines and processes (Adler & Obstfeld, 2007;Baumard & Starbuck, 2005). Returns to invest-ments that exploit existing capabilities are imme-diate, whereas returns to exploration and learningare distant and uncertain (March, 1991). Instead offacing uncertainty, organizational actors often failto look into the distant future, choosing to “solvepressing problems rather than develop long-runstrategies” (Cyert & March, 1963: 119). In essence,existing routines can become sources of inertia(Edmondson, Bohmer, & Pisano, 2001; Leonard-Barton, 1992; Levitt & March, 1988).

Organizational theorists have proposed anumber of approaches for encouraging explora-tion and discontinuous advances in learning (for

For their helpful comments on previous versions of thismanuscript, we thank Bettina Buechel, Rich Burton, LauraCardinal, Christina Fang, John Joseph, Theresa Lant, RickLarrick, Ed Locke, Frances Milliken, Elizabeth Morrison,Gerardo Okhuysen, Randall Peterson, and three anonymousreviewers. We also thank participants in the Harvard–MITOrganizational Economics Seminar Series and the 2008Academy of Management annual meeting.

� Academy of Management Review2011, Vol. 36, No. 3, 544–566.

544Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyrightholder’s express written permission. Users may print, download, or email articles for individual use only.

reviews see Greve, 2003; Huber, 1991; Levitt &March, 1988; Sitkin, Sutcliffe, & Weick, 1998).March (1991), for example, highlighted the valueof turnover that creates new perspectives. Chris-tensen (1997) argued for forming new operatingunits that are not encumbered by existing rou-tines and capabilities. Siggelkow and Levinthal(2003) considered the role of decentralization.Levinthal and March (1993) suggested a numberof diverse tactics for encouraging change, in-cluding leveraging employees who have failedto thrive within the existing order or explicitlymanipulating risk preferences in the organiza-tion. These ideas for promoting exploration havebeen useful but also have nontrivial drawbacks,such as the substantial structural changes re-quired to implement new operating units or theloss of still valuable tacit knowledge throughturnover. Moreover, these and other ideas forpromoting exploration often have been built ona foundation of general or abstract theoreticalreasoning, leaving the underlying mechanismsunderspecified. All of these issues, as well asthe inherent complexity of the topic, havehelped to keep exploration at the forefront ofdebates and study within the organizational re-search community (for recent commentaries seeFang & Levinthal, 2009; Kim & Rhee, 2009; Raisch& Birkinshaw, 2008).

Both theory and intuition suggest that meth-ods for promoting exploration and changeshould facilitate attention, energy, and action inthe domain of alternative routines and capabil-ities. Without attention being channeled to al-ternative futures, new paths are not likely to beconsidered (e.g., D’Aveni & MacMillan, 1990;Drazin & Sandelands, 1992; Starbuck, 1983).Without energy and enthusiasm for majorchange, challenges to the status quo are un-likely (e.g., Beer, Eisenstat, & Spector, 1990).Without coordinated action, trial-and-error ex-perimentation is less likely to yield meaningfulresults (e.g., Huber, 1991; Sitkin et al., 1994). Theattention-based view of the firm (e.g., Ocasio,1997), research into organizational change andadaptation (e.g., Barnett & Pratt, 2000; Kotter,2008), and studies of organizational learningand design (e.g., Argyris, 1985) directly highlightthe importance of these cognitive, affective, andbehavioral mechanisms, respectively.

One method for exploration that might insti-gate the mechanisms discussed above is the useof seemingly impossible organizational goals—

commonly referred to as stretch goals. Becausethey are extreme, stretch goals have been ar-gued to serve as jolting events that disrupt com-placency and promote new ways of thinking andacting (e.g., Hamel & Prahalad, 1993; Rousseau,1997). In effect, the imposition of such extremegoals can be similar to an autogenic (i.e., inten-tional, internally generated) crisis meant to spurchange (e.g., Barnett & Pratt, 2000; D’Aveni &MacMillan, 1990). By forcing a substantial eleva-tion in collective aspirations, stretch goals canshift attention to possible new futures and per-haps spark increased energy in the organiza-tion. They thus can prompt exploratory learningthrough experimentation, innovation, broadsearch, or playfulness as organizational actorsseek new or varied approaches to reach the tar-get. Stretch goals also can enhance tangibleperformance outcomes as gaps between aspira-tion and current performance elicit and guideeffort and persistence (Cyert & March, 1963).

Reflecting on both performance and explora-tion, Rousseau put it this way: “[Stretch goals]motivate high performance by mandating cre-ativity and assumption-breaking thinking”(1997: 528). Winter said the following in the con-text of stretch goals and organizations develop-ing new capabilities: “It is not a secret that highaspirations can often contribute to high achieve-ment” (2000: 990). These sentiments are echoedin the business press. Steve Kerr, former ChiefLearning Officer at General Electric and Gold-man Sachs, illustrated the popular motive for,and the method of, stretch goals:

If done right, a stretch target . . . gets your peopleto perform in ways they never imagined possible.It’s a goal that, by definition, you don’t know howto reach. You might, for instance, ask people tocut costs by half or reduce product-developmenttime from years to months . . . [in order to] finddramatically new ways of doing business (Sher-man, 1995: 231).

Pursuing goals that are seemingly impossiblemight stimulate exploratory learning specifi-cally because radically new approaches are re-quired. Southwest Airlines and Toyota provideexamples. After being forced to sell part of itsfleet early in its history, Southwest Airlines set agoal of ten-minute turnaround times at airportgates in an attempt to use its few remainingplanes more efficiently. Although officials withthe U.S. Federal Aviation Administration, Boe-ing, and competing airlines believed the goal to

2011 545Sitkin, See, Miller, Lawless, and Carton

be unachievable (as did many Southwest em-ployees), the seemingly impossible ten-minuteturnarounds were ultimately accomplished byemploying an approach (drawn from race car pitcrews) that was radically new and unfamiliar tothe airline industry (Freiberg & Freiberg, 1996).Likewise, some have argued that at Toyota theseemingly impossible goal of 100 percent near-term improvement in fuel efficiency (relative toprevailing standards) played a key role in thedevelopment of hybrid vehicle technology (Takeu-chi, Osono, & Shimizu, 2008). Beyond these twoanecdotal illustrations, researchers have sug-gested that the use of stretch goals remains fairlycommon in practice (for examples see Collins &Porras, 1994; Takeuchi et al., 2008; Thompson,Hochwarter, & Mathys, 1997).

Goals and aspirations in general have longplayed an important role in organization the-ory (e.g., Cyert & March, 1963), and stretchgoals in particular have generated interestamong the business press and the variousscholars noted earlier. Yet pushing the bound-aries of goal attainability raises organiza-tional implications that require deeper analy-sis. In this article we integrate the literatureon organizational goals with research onlearning to examine the role of seemingly im-possible goals in facilitating exploratorylearning while promoting, or at least not sac-rificing, performance. Specifically, we con-sider a number of fundamental questions atthe organizational level of analysis: What arethe mechanisms through which this class ofgoals might promote exploratory learning andorganizational performance? What are therisks to such extreme goals that could nega-tively affect the organization’s capacity to per-form and to learn? Do stretch goals increaselearning or performance in some circum-stances but decrease them in others? Do orga-nizations that could benefit the most fromstretch goals exhibit the greatest propensity topursue them?

In gaining an understanding of why organiza-tions would use seemingly impossible goalsand how their use influences exploratory learn-ing and performance, we address an importanttheoretical gap at the intersection of research onorganizational goals and organizational learn-ing processes. In particular, we contribute newinsights on how the use of extreme goals mightrelate to exploring new and innovative practices

(e.g., Levinthal & March, 1993; March, 1991; Rah-mandad, 2008; Uotila, Maula, Keil, & Zahra,2009), organizational risk taking (e.g., Singh,1986; Sitkin & Pablo, 1992), learning under am-biguous feedback (March & Olsen, 1976; Weick,1979, 1995), and dynamic capability develop-ment (e.g., Agarwal & Helfat, 2009; Capron &Mitchell, 2009; Winter, 2000). Our work revealsways that the extremity of stretch goals mightchallenge core assumptions about the relation-ships among aspirations, learning, and firm per-formance (e.g., Ethiraj & Levinthal, 2009; Cyert &March, 1963; Greve, 1998; Lant, 1992; Lant & Sha-pira, 2008; Mezias, 1988). We suggest that asgoals become extreme, there are complex yetpredictable organizational effects that are likelyto be negative except under a limited set ofspecifiable circumstances. Moreover, we con-tribute to scholarly thinking on organizationalchange and adaptation by examining why orga-nizations would be drawn to using stretch goalsas intentional, internally generated jolts or crises(e.g., Barnett & Pratt, 2000; D’Aveni & MacMillan,1990), as well as the conditions under which suchjolts might successfully or unsuccessfully triggerdiscontinuous advances in learning (e.g., Beer etal, 1990; Meyer, 1982; Romanelli & Tushman, 1994).Our theoretical analysis also advances the under-standing of the little-known effects of unattain-able goals (e.g., Garland, 1982, 1983; Locke, 1982,2004; Rousseau, 1997) by examining the underly-ing mechanisms through which such goals canlead to collective outcomes.

In the following section we specify our con-ceptual space by providing a definition ofstretch goals. We then examine the underlyingcognitive, affective, and behavioral mecha-nisms through which stretch goals might posi-tively or negatively influence organizationallearning and performance outcomes. Building di-rectly on this discussion, we formulate pro-positions around recent performance and slackresources as the key contingency factors deter-mining when stretch goals will facilitate versusdisrupt learning and performance. We follow withpropositions concerning how these same contin-gency factors also determine the likelihood thatan organization will be drawn to using stretchgoals. As part of our closing discussion, we recon-sider reported stretch goal success stories in lightof our analysis and highlight important contextsfor future empirical inquiry.

546 JulyAcademy of Management Review

STRETCH GOALS DEFINED

Consistent with an early stage of conceptualdevelopment, stretch goals have not been pre-cisely defined, and the term has not been usedconsistently within or across previous commen-taries. Even so, earlier use of the term does pro-vide guidance in formulating a definition. Draw-ing on prior descriptions (e.g., Collins & Porras,1994; Hamel & Prahalad, 1993; Rousseau, 1997;Sherman, 1995), we define a stretch goal as anorganizational goal with an objective probabil-ity of attainment that may be unknown but isseemingly impossible given current capabilities(i.e., current practices, skills, and knowledge).

Because we define stretch goals in terms of anunknown yet seemingly impossible (i.e., 0 per-cent) probability of attainment, we depart mark-edly from the focus in organizational behaviorresearch on challenging goals, which have anonzero (typically, 10 percent) probability of at-tainment (Locke & Latham, 1990). We do not de-part, however, from the usual focus on outcome-oriented performance goals as opposed toprocess-oriented learning goals (e.g., Seijts &Latham, 2005; Winters & Latham, 1996). Setting aperformance goal entails specifying a tangibleoutcome to be reached (usually a quantifiablelevel of performance), such as a specific reduc-tion in turnaround times at airport gates, a per-centage increase in fuel efficiency of vehicles, areduction in cycle time in a production process,or an increase in sales from new products. Oncethat performance goal is set, the entity chargedwith pursuing it must devise strategies to reachthe targeted outcome. Although learning can re-sult as a by-product of dealing with the de-mands of meeting a performance goal, it is im-portant to emphasize that stretch goals are stillarticulated by the organization in terms of aspecific level of performance or output. In con-trast, setting a learning goal that does not makeexplicit a specific outcome to be reached (but,rather, articulates the acquisition of proceduralknowledge as the end in itself) would not qual-ify as a stretch goal under our definition.

Our characterization of stretch goals high-lights that they differ from ordinary difficultgoals in two important respects: (1) extreme dif-ficulty—an extremely high level of difficultythat renders the goal seemingly impossiblegiven current situational characteristics and re-sources—and (2) extreme novelty—there are no

known paths for achieving the goal given cur-rent capabilities (i.e., current practices, skills,and knowledge). Although these dimensions im-ply each other, extreme difficulty and noveltystress different aspects of the stretch goal con-struct (the specified performance outcome andknowledge of the means to reach it, respec-tively) and also directly relate to the two differ-ent core outcomes of interest in this article (or-ganizational performance and organizationallearning, respectively). Thus, below we elabo-rate on the dimensions separately in order tofacilitate systematic theorizing about distinctstretch goal effects.

Extreme Difficulty

Stretch goals involve extreme or radical ex-pectations. Even if the stretch goal involves en-hancing a process already in place (e.g., theelapsed time between an order and delivery of aproduct to the customer), the desired processimprovement extends beyond what is possiblewith current capabilities. Returning to theSouthwest Airlines example, the goal of a ten-minute turnaround certainly involved a familiartask (efficient turnarounds at airports), but thetarget was dramatically beyond the company’sthen-current performance limits, as well as theindustry average and range of industry prac-tices. Regardless of the method Southwest Air-lines deployed to try to reach the stretch target,attainment would at best be extraordinarily dif-ficult and was perceived to be impossible.

We note that a goal need not be universally(or eternally) impossible in order to qualify as a“stretch” for an organization in a given context.Although it is likely that seemingly impossiblegoals for one organization are also seeminglyimpossible for most other organizations in theindustry at the time, there could be exceptions.Overall, the determination of extreme difficultyis context specific.

Extreme Novelty

A second aspect of stretch goals is the lack ofa discernible path to attainment. When an orga-nization lacks the skills, knowledge, or practicesto attain a stretch goal and does not have knowl-edge of any feasible approaches, it is effectivelyforced to search outside of its normal routinesand knowledge to see if any ways to achieve the


goal exist or can be created. At issue are theparticular circumstances of the organizationcharged with pursuing the goal. To continuewith the Southwest Airlines illustration, the goalof a ten-minute turnaround could only beachieved by learning to employ a radically newapproach. There were no guiding templates touse within the organization or even within theairline industry. In other words, the path toachievement was unimaginable at the time thegoal was set, and, thus, novel means were re-quired to achieve the goal.

STRETCH GOALS AS FACILITATORS ANDDISRUPTORS OF ORGANIZATIONAL

LEARNING AND PERFORMANCE

There is as yet no explicit theory of the poten-tial effects of seemingly impossible goals onexploratory learning and organizational perfor-mance. Indeed, macrolevel goals are often in-voked in the organizational literature, but theirunique effects have rarely been examined, andunattainable goals remain outside nearly all ofthe available scholarly work. Much of the pub-lished work focused directly on organizationalstretch goals has been restricted to commentar-ies or applied case studies (Collins & Porras,1994, 1996; Golovin, 1997; Hamel & Prahalad,1993; Hughes, 2001; Kerr & Landauer, 2004; Rous-seau, 1997; Sherman, 1995; Takeuchi et al., 2008;Thompson et al., 1997). With few exceptions (e.g.,Hughes, 2001; Kerr & Landauer, 2004; Sherman,1995), these authors have focused solely on thepresumed benefits of seemingly impossiblegoals, and their work has remained general,rarely proposing specific causal mechanisms.

Notwithstanding these limitations, we havebeen able to draw on work in organizationaltheory and cognition to analyze how stretch goalusage can influence learning and performanceoutcomes through a variety of specific mecha-nisms that we sort as cognitive, affective, orbehavioral (as summarized in Figure 1). Ouranalysis surfaces both potential facilitative anddisruptive stretch goal effects. We then proposea set of contingency factors that determinewhether stretch goals will have positive or neg-ative effects on learning and performance.

In line with our central purpose and with priorwork, we conceptualize learning as explorationthat yields meaningful insights regarding newpractices and capabilities (e.g., March, 1991). We

conceptualize performance in terms of an orga-nization’s tangible outcomes along such dimen-sions as productivity, product or service quality,profitability, growth, and market share (e.g., seeHitt, 1988, and Quinn & Rohrbaugh, 1983).

Facilitative Effects of Stretch Goals onExploratory Learning and Performance

Facilitative effects via cognition. Attention isa well-recognized mechanism of influence onorganizational learning and performance out-comes (Cyert & March, 1963; Ocasio, 1997; Sut-cliffe, 1994). Because stretch goals involve ex-treme redefinitions of what an organization iscapable of being or achieving, they can capture,shift, and refocus attention. According to Rous-seau, “Where performance expectations are el-evated well beyond the limits of past experi-ence” and “where previously successfulframeworks are questioned, revised, or dis-carded, prior experience is often a poor guide forstretch-goal achievement . . . [and this] shifts theperformers’ attention away from old routinesand assumptions toward novel and creative ap-proaches” (1997: 529). Stretch goals present aninformation-processing challenge (Galbraith,1973) because the organization needs to findnew sources and types of information and alsonew ways to process that information. A positivereaction to this challenge would be for the orga-nization to become heedful (Weick, 1995; Weick,Sutcliffe, & Obstfeld, 2005) or vigilant (Janis &Mann, 1977) in proactively scanning and assess-ing the situation. The organization could thusbecome more open to new information from avariety of sources (Huber, 1991; O’Reilly, 1982;Sutcliffe, 1994) and question the validity of oldassumptions, old information, and old frame-works (Meyer, 1982; Rousseau, 1997). These con-ditions can facilitate learning processes.

Because stretch goals move an organizationinto uncharted waters, they naturally requireflexible thinking about revisable alternativestrategies for goal attainment (March, 1991;March & Olsen, 1976). Thus, seemingly impossi-ble goals can elicit a more open, reflective, andopportunity-oriented focus that is beneficial forperformance (Sitkin, 1992). By allowing the orga-nization to more freely tap previously underuti-lized sources of information and insight, the joltof stretch goals can cause an organization tomore avidly attend to and consider novel paths


for pursuing the stretch target (March, 1991). Aspart of this response, capabilities could be care-fully reevaluated for their potential to be recom-bined in novel ways (e.g., Henderson & Clark,1990), which would focus attention productivelyon controllable internal resources and wouldhave positive consequences for performance(Beer et al., 1990; Lant & Shapira, 2008; Levinthal& March, 1981).

Facilitative effects via affect. Seemingly im-possible goals can facilitate organizationallearning and performance by positively influ-encing the collective emotion and initiative in-fused into the organization (Adler & Obstfeld,2007). This is not to suggest that organizations

feel emotions but, rather, to recognize that sig-nificant opportunities or threats can have a col-lective impact that goes beyond individualmembers’ affective responses through conta-gion processes (Barsade, 2002; Barsade & Gib-son, 2007; Bartel & Saavedra, 2000).

Because stretch goals involve such high lev-els of ambition directed toward novel and unfa-miliar opportunities, they can generate energyand greater initiative to learn by evoking arange of positive collective reactions, such asoptimism, urgency, enthusiasm, curiosity, andplayfulness (March, 1976, 1991). When U.S. Pres-ident John Kennedy announced in 1961 the goalof landing a man on the moon within a decade,

FIGURE 1Mechanisms Through Which Stretch Goals Can Influence

Organizational Learning and Performance

Facilitation Disruption

Learning

Cognitive Vigilance (heedfulness/mindfulness) Systematic processing of new information Openness

Affective Enthusiasm, energy Optimism Sense of urgency Curiosity, playfulness

Behavioral Trial-and-error cycles Broad search for new sources and discontinuous advances

Performance

Cognitive Focus on internal/controllable factors Opportunity interpretations Attention on usable new information sources and analyses

Affective Initiative to improve High resilience to negative feedback

Behavioral Effort and persistence Effective strategy selection

Cognitive Hypervigilance Inability to process new information

Affective Fear Helplessness Aversion to change Defensiveness

Behavioral Chaotic change Insufficient familiarity for interpreting feedback

Cognitive Focus on external/uncontrollable factors Threat interpretations Attention on quick fixes

Affective Low commitment to goal Low resilience to negative feedback

Behavioral Threat rigidity Impaired coordination Resource diversion resulting in loss of beneficial routines


he introduced a target that excited those he led,even though at the time they had no idea how toachieve the goal, or even if its achievement waspossible. More generally, stretch goals can stim-ulate exploration by highlighting a potentiallybetter and more exciting future (cf. Shamir,House, & Arthur, 1993). By creating a sense ofurgency to take on new challenges and to movetoward new understandings, stretch goals canimpose a “crisis” that stimulates the initiativewithin or across the organization to undertakeexperimentation and learning (Barnett & Pratt,2000; Baumard & Starbuck, 2005; D’Aveni & Mac-Millan, 1990; Kim, 1998). In other words, the cou-pling of positive affective drivers (e.g., enthusi-asm, energy) with modest stress-related drivers(e.g., urgency) can provide the spark and collec-tive initiative to explore and learn (Argyris &Schon, 1978; Kotter, 2008).

The extreme difficulty of attaining stretch goalsshould also elevate aspirations (Rousseau, 1997),which is critical for ensuring that organizationsstrive to increase performance. Organizational the-ory suggests that aspirations well above existingperformance levels create a problem to be solvedand increase the initiative and desire to find a solu-tion (Cyert & March, 1963; Greve, 1998; Lant, 1992;Levinthal & March, 1981). In Greve’s work (1998), forexample, aspirations well above existing perfor-mance led to enhanced motivation to improve per-formance. Strategic planning research also sug-gests that plans involving goals positively affect thedesire to enhance performance, because strategicgoal setting gives direction to the firm and promotesadaptive thinking (e.g., Miller & Cardinal, 1994). Theliterature on charismatic leadership, visionary lead-ership, and transformational leadership further sug-gests that the establishment of targets beyond cur-rent capabilities builds collective enthusiasm andresilience (Bass & Riggio, 2005; Dvir, Eden, Avolio, &Shamir, 2002; Nanus, 1992; Shamir et al., 1993).

Facilitative effects via behavior. Since there isno known path to achieving a stretch goal, put-ting such a goal in place may generate actionsdirected toward exploring and learning newpractices. When compared with other condi-tions, the extreme demands of stretch goalscould stimulate broader and more active searchfor ideas and solutions (e.g., Baum & Dahlin,2007; Fang & Levinthal, 2009; Mezias, 1988).Stretch goals could also propel the search formore radical, discontinuous advances (Cyert &March, 1963; Raisch & Birkinshaw, 2008), involv-

ing actions such as making contacts with unfa-miliar sources of ideas, imitating innovativetechniques (e.g., Sitkin et al., 1994), or undertak-ing trial-and-error learning (see Ingram & Baum,1997, and Levitt & March, 1988). Moreover, byloosening old restraints and providing focusand energy, stretch goals may allow for trial-and-error learning to be undertaken with fastercycle times (Argote, 1999; Argyris, 1985; March,1991). In effect, stretch goals can prompt searchand experimentation behaviors traditionally as-sociated with learning.

Stretch goals can also instigate actions thathave positive effects on performance. Once theorganization has identified and focused on astretch target, prior research suggests that itsuse of specific and measurable (as opposed tovague) goals can lead to increases in organiza-tional performance outcomes via greater collec-tive effort, persistence, and the development ofcoherent action strategies for attaining the tar-get (e.g., Chesney & Locke, 1991; Cyert & March,1963; Smith, Locke, & Barry, 1990). For example,Chesney and Locke (1991) found that specificallyarticulated goals influenced the development ofbusiness strategies that increased simulatedfirm performance in an experiential strategicmanagement exercise. Moreover, research ontop management teams has connected upper-echelon agreement on the goals of the firm tofinancial performance (e.g., Colbert, Kristof-Brown, Bradley, & Barrick, 2008; Dess, 1987). Mov-ing from the use of attainable goals to the caseof stretch goals, to the extent that top manage-ment agreement on a stretch goal signals agreater sense of specificity and unified purpose,more effective pathways and strategies for ap-proaching the goal could lead to higher perfor-mance. Essentially, when a stretch goal is artic-ulated as achieving a specific outcome, parallelor independent strategies can be more easilydeveloped and coordinated as the beacon-likestretch target directs and channels action to-ward a singular end point.

Disruptive Effects of Stretch Goals onExploratory Learning and Performance

Disruptive effects via cognition. In contrastto the positive dynamics of vigilance or mind-fulness discussed above (Janis & Mann, 1977;Weick, 1995), the jolt of stretch goals couldconceivably elicit negative attentional re-


sponses that compromise information process-ing and the capacity to learn. In the face of aseemingly impossible problem that is not wellunderstood and for which there are no discretesteps that constitute credible paths toward solu-tion, the organization may respond to the over-whelming situation with hypervigilance (Janis &Mann, 1977). With such a response, the organi-zation’s information processing becomes disor-ganized, impulsive, and less systematic in theconsideration of alternatives. Thus, instead ofpresenting a stimulating intellectual challengefor information processing, stretch goals couldconstitute an information processing “bridge toofar,” in that the organization simply lacks thecapacity to match the complexity and demandsof its situation or to incorporate new approachesor inputs that could lead to learning (Weick,1979).

The lack of obvious capabilities to respond tothe demands of stretch goals can also shift at-tention to potential quick fixes from outside(Beer et al., 1990), even though the organizationmay not have the absorptive capacity to importthese approaches (Cohen & Levinthal, 1990;D’Aveni & MacMillan, 1990). Attending to outsideideas also diverts attention from a more produc-tive focus on internal resources and ideas forchange (Beer et al., 1990; Lant & Shapira, 2008).While shifting some attention externally is gen-erally positive for exploration, the extreme dif-ficulty of pursuing a stretch goal can lead theorganization to focus attention too much or toohaphazardly on outside ideas (e.g., obsessivemonitoring or unsystematic benchmarking ofother firms), which would directly limit perfor-mance as the organization directs insufficientattention to opportunities that are most usefulgiven its unique history and environment (Cyert& March, 1963).

Disruptive effects via affect. The contagionprocesses that drive collective affect in organi-zations occur regardless of whether the affectiveresponse is positive or negative (Barsade, 2002).While the high levels of ambition associatedwith stretch goals could raise the level of posi-tive collective affect, as discussed earlier,stretch goals signal the need for change and,thus, simultaneously have the potential to elicitthe negative affective responses that often ac-company change (Barnett & Pratt, 2000). In par-ticular, facing a goal that is seemingly impossi-ble could instigate a negative affect contagion

whereby exploration and willingness to try newapproaches are impeded by a sense of collectivefear, helplessness, and demotivation (Sitkin,1992). Under such conditions, any incentives toexplore induced by stretch goals would bedwarfed by the more salient threat created bythem. A high level of negative collective affect isgenerally expected to constrict the motivation touse learning-enhancing processes and the ca-pacity to absorb learning opportunities that mayserendipitously present themselves (Cohen &Levinthal, 1990; Levinthal & March, 1993; Levitt &March, 1988; Woodman, Sawyer, & Griffin, 1993).

Stretch goals can also have disruptive effectson performance if collective commitment to thegoal becomes compromised by widely held em-ployee perceptions that a seemingly impossiblegoal is unrealistic or unworthy of pursuit. Theextremity of stretch goals can also dampen col-lective affective responses like satisfaction andmorale, because organizational attempts toreach the goals are likely to involve one or morefailures. Thus, less extreme goals that are diffi-cult but still attainable are often advised (Locke& Latham, 1994). Indeed, using goals that aredifficult enough to require great effort or persis-tence, yet are still within reach, is consistentwith many motivational perspectives (e.g., At-kinson, 1964; Heath, Larrick, & Wu, 1999; Locke &Latham, 1990; Vroom, 1964). For example, Heathet al. (1999) built on prospect theory to contendthat because a goal serves as a reference point,initiative for working toward the goal dimin-ishes with distance from the goal. Taken to-gether, stretch goals have the potential to dis-rupt performance by diminishing collectiveinitiative, morale, commitment, and resiliencein the face of setbacks.

Disruptive effects via behavior. By definition,stretch goals are associated with extreme tar-gets whereby current capabilities are not ade-quate in any obvious way for goal attainment.As such, actions associated with stretch goalimplementation could create too large a breakfrom the past, or the actions may simply lead totoo many simultaneous changes to effectivelydistinguish signal from noise (March, 1976). Ineither case, organizational decision makerswould be unable to learn which enactedchanges are effective and which are ineffectiveor detrimental (March & Olsen, 1976). Moreover,organizational learning is best facilitatedwithin at least moderately familiar territory


where there is a frame of reference to under-stand causal relationships and interpret feed-back from changes that are made (Cohen &Levinthal, 1990; Sitkin, 1992; Weick, 1984). With-out some degree of familiarity, organizationslack an effective frame, and feedback from ac-tions and change efforts is experienced asambiguous (cf. Levitt & March, 1988; March &Olsen, 1976). The novel means associated withstretch goal pursuit imply that there is little “do-main relevance” to bring to bear on the situa-tion, since “it is hard to be intelligent about thatwhich is unfamiliar” (Sitkin, 1992: 245). Thus,conditions that could facilitate effective learn-ing, such as controllable change in familiar ter-ritory with interpretable feedback, are unlikelyto hold in the presence of stretch goals.

The extreme novelty of stretch goals couldgenerate actions that complicate organizationalperformance. Because it is not clear initiallyhow to start working toward a seemingly impos-sible goal, the inherent novelty and task com-plexity involved in attaining a stretch goal cancompromise some types of organizational per-formance (e.g., Carley & Lin, 1997; Perrow, 1979).Moreover, an additional concern is the magni-tude of change that would typically be requiredin attempting to bridge the gap between currentcapabilities and the attainment of a stretchgoal. Greater scale change can be disruptive toperformance because problems encounteredand lessons learned are often not amenable toeffective coping (Beer et al., 1990), whereas re-duced scale of change creates more manage-able problems with smaller performance mile-stones (Sitkin, 1992; Weick, 1984). The prospect oflarge-scale change induced by stretch goals canalso instigate a threat-rigidity response (Staw,Sandelands, & Dutton, 1981)—an inability to actwhen under threat—that can undermine perfor-mance. Threat rigidity would be particularlymaladaptive in the context of stretch goal pur-suit, since rigidly clinging to existing practicescannot result in stretch goal attainment.

The extreme novelty and difficulty of attain-ing stretch goals could also directly impair co-ordination. Mandating the use of an organiza-tional stretch goal can be done almost instantly,but actual attempts at implementation demandradical organizational change under vagueguidance. Urgent change under such ambigu-ous circumstances can cause different organiza-tional subunits to develop discrepant views

about causality and appropriate action (Beer etal., 1990; Levitt & March, 1988). The resultingplurality of “stories” ascribed to the actions andoutcomes associated with attempting to reachthe stretch goal is likely to impair coordinationamong subunits that have separate purposes,specialties, and interests (Lawrence & Lorsch,1967; Martin, 1992). Information distribution andintegration, which are fundamental coordina-tion elements inherent in organizational perfor-mance (as well as learning), are likely to benegatively affected as a result (Huber, 1991).

Finally, because stretch goals require newroutines and resources, working to attain themcould reduce (or eventually eliminate) the effec-tive organizational exploitation of existing ca-pabilities, practices, and investments that areotherwise working well. Exploring new routinesis, of course, a large part of the rationale andpromise of using stretch goals. But diverting toomuch or diverting for too long from existing,well-functioning investments can undermineoverall performance (March, 1991; Perrow, 1979),even in the best case scenario where the orga-nization is simultaneously reaping some learn-ing and performance returns from the stretchpursuit. That is, while stretch goals are intendedto break problematic inertial forces, beneficialpath dependencies can also be lost if any rou-tines that remain useful are thrown out. The joltthat stretch goals provide can thus outstrip theorganization’s existing capabilities andresources.

Our analysis in this section has suggested anumber of specific mechanisms by whichstretch goals could have either facilitative ordisruptive effects, as summarized in Figure 1. Inthe following section we propose a contingencymodel of the primary organizational factors thatultimately predict whether stretch goals will fa-cilitate or impede exploratory learning andperformance.

CONTINGENCY FACTORS DETERMININGSTRETCH GOAL EFFECTS ON LEARNING

AND PERFORMANCE

To consider the contingencies that influencewhen organizations benefit or suffer from theuse of stretch goals, we focus on the two coreorganizational factors that reflect the system’scapacity to extract value from radical changesor new approaches: recent performance and


slack resources (Cyert & March, 1963). In termsof the mechanisms we analyzed in the priorsection, we seek to answer the question, “Howdo recent performance and level of slack re-sources influence the balance between the fa-cilitative and disruptive aspects of the cogni-tive, affective, and behavioral mechanisms bywhich stretch goals can influence exploratorylearning and performance?” In other words,we propose a contingency framework to ex-plain when stretch goals will have positive ornegative effects on learning and performanceoutcomes depending on the levels of recentperformance and slack resources of the orga-nization using such goals.

Recent Performance

Is an organization fresh from success (e.g.,recent performance above a benchmark, suchas the organization’s own performance in aprior period, average industry performance, orperformance of a particular competitor) better sit-uated to leverage the potential benefits of stretchgoal use and to withstand fallout from any fail-ure? With respect to the mechanisms summarizedin Figure 1, there are reasons to expect that stron-ger recent performers, if they use stretch goals, arewell situated to achieve facilitative effects on bothlearning and performance outcomes, whereasweaker recent performers are situated for disrup-tive effects from the use of stretch goals.

Recent performance is likely to drivewhether stretch goals shift organizational at-tention in a way that is focused productively orcounterproductively. Having just experiencedsuccess, stronger recent performers are lesslikely to perceive an immediate threat; thus, ifthey undertake a stretch goal, they should bemore open to new ideas and more mindful inthe scanning and processing of new informa-tion, both of which foster learning (e.g., Levitt& March, 1988; Weick et al., 2005). As part ofbeing more mindful and vigilant, stronger re-cent performers are more likely to attend tocontextual features and organizational capa-bilities that are internal and controllable (e.g.,Staw et al., 1981; Sutcliffe, 1994), precisely be-cause they have a (recent) history of deployingthose features successfully (March & Olsen,1976). Thus, in striving to reach a stretch goal,any routines that could be leveraged for radi-cal performance-enhancing recombinations

would be cognitively available to, and de-tected by, stronger recent performers.

Weaker recent performers, however, do nothave the attentional advantages afforded bysuccess. Because they are already in a com-promised state owing to recent losses, ifweaker recent performers choose to pursue aseemingly impossible goal, they are morelikely to be overwhelmed by its demands andresort to hypervigilant, disorganized, or evenfrantic information processing that is disrup-tive to learning. Scattered attention and un-systematic information processing are alsodisruptive to performance such that weakerrecent performers may attempt to reach astretch goal by haphazardly or obsessivelymonitoring others in the industry for exter-nally sourced quick fixes. Such external re-sources and practices may not pertain to in-ternal resources (Beer et al., 1990), leverageexisting strengths (March & Olsen, 1976), orallow for forging new paths and solutionsthrough recombination (Henderson & Clark,1990).1

Recent performance would also be expectedto influence whether the collective affective re-sponses to stretch goals facilitate or disruptlearning and performance. Success generatesoptimism, enthusiasm, and commitment and be-comes self-reinforcing (March & Olsen, 1976;Weick, 1984). Thus, if strong recent performersuse stretch goals, they should be more able toleverage optimism and energy from their recentsuccess toward learning solutions for goal at-tainment and increased performance. In con-trast, poor recent performance underminesmany of these affective mechanisms. Poor re-cent performers attempting to reach a stretchgoal are thus more susceptible to collective re-

1 We note one situation at the far extreme of poor perfor-mance, where setting stretch goals is the only option forsurvival. Like the “Hail Mary” pass at the end of a U.S.football game, the pursuit of stretch goals when an organi-zation has no other option could be normatively superior solong as it does not hasten death. However, one could alsodraw an analogy to the pursuit of a miracle cure for aterminal disease, which introduces the notion that if settingstretch goals consumes fungible resources that could still beused more effectively in other pursuits (e.g., employees de-voting time to find other employment or more resourcesavailable to pay creditors), then the use of stretch goals evenfor survival may not always be the superior option to a quickand relatively painless organizational death.


sponses of fear or defensiveness that would ob-struct the readiness to try new things and learnfrom them (Argyris, 1985; Sitkin & Pablo, 1992), aswell as hamper the initiative and commitmentneeded to effectively pursue the performanceopportunities offered by a stretch goal (Baumard& Starbuck, 2005).

Finally, the actions taken in the attempts toreach a stretch target would conceivably be verydifferent as a function of an organization’s re-cent performance and would therefore bringabout different learning and performance out-comes. To the extent that success breeds thepersistent utilization of the routines and prac-tices that led to that success (Audia, Locke, &Smith, 2000; March & Olsen, 1976; Weick, 1984),strong recent performance might initially leadthe organization down the erroneous path of try-ing to reach a stretch goal only by repeatingactions undertaken in the past. However, once itdetermines that such an approach is not suffi-cient for reaching a stretch target, a strongerrecent performer, lacking an existing threat,could respond to the crisis imposed by stretchgoals with greater flexibility (Barnett & Pratt,2000), such as by devising new strategies tomeet the target and undertaking more effortfulsearch activities, which are actions that canyield both learning and performance benefits.

Poor recent performance, however, can com-promise the capacity to undertake systematicand controlled experimentation, thus leaving or-ganizational actors to approach a stretch goalby instituting more chaotic changes that will notprovide the clear feedback needed for completelearning. Moreover, recent performance prob-lems signal that an organization is already un-der threat. It may even be the case that thestretch goal itself is related to the performanceproblems. Thus, encountering a stretch goalwhen there is existing pressure and where it isnot immediately clear how to approach the goalcould evoke a dysfunctional threat-rigidity re-sponse (Barnett & Pratt, 2000; Staw et al., 1981).Overall, stronger recent performers usingstretch goals are better situated to experiencefacilitative effects on learning and performance,whereas weaker recent performers using stretchgoals are situated for disruptive effects.

Proposition 1: For organizations withstrong (weak) recent performance, theuse of stretch goals will yield positive

(negative) effects on learning andperformance.

Slack Resources

The presence of financial or other resourcesthat have not been committed or deployed in thesystem (Bourgeois, 1981) and are available forthe discretionary use of management (Nativi-dad, 2009) creates a buffer that scholars haveconceptualized as unabsorbed slack. How mightgreater unabsorbed slack resources affect thecapacity for organizations to reap successfuloutcomes when pursuing the seemingly impos-sible? Existing research suggests that greaterslack resources protect against many of the dis-ruptive mechanisms listed in Figure 1 andthereby help to facilitate positive effects on bothexploratory learning and performance for orga-nizations using stretch goals.

Slack serves a practical role in that findingexciting opportunities and identifying and cul-tivating internal capabilities to reach a seem-ingly impossible goal will typically demand theavailability of people, money, time, and otherresources. As a result, greater slack imbues anorganization that opts to use stretch goals withcertain cognitive advantages. By providing thetime and access to the resources necessary todiscover positive potential outcomes and attain-ment paths, slack allows for more openness tonew information and vigilance in the processingof that information, which is beneficial forlearning.

Moreover, the presence of slack resourcesshould shift attention in a way that could bebeneficial for performance (Ferrier, 2001; Nohria& Gulati, 1996; Young, Smith, & Grimm, 1996). Weargue that, for an organization that is pursuing astretch goal, having sufficient slack resourcescreates a situation in which the stretch goal ismore likely to be interpreted as an opportunityto generate and sift through new ideas that arepotentially usable based on internal existingcapabilities. In contrast, limited slack would ob-struct attention to possibly innovative (yet notobvious) leveraging of current capabilities,since detecting novel recombinations takes timeand effort that would be less available.

Having slack also enables stretch goals tofacilitate learning and performance by servingas both a literal and psychological bufferagainst the potentially negative collective affec-


tive responses to seemingly impossible goals.Because stretch goals involve the prospect ofextreme change without an obvious method forgoal attainment, collective responses of fearand helplessness may ensue if there are insuffi-cient available resources to bring to bear on thesituation. Greater slack reduces the pressure ofobtaining resources necessary for trying to reacha seemingly impossible goal, and, thus, the pur-suit of stretch goals can be met with more playful-ness, curiosity, and enthusiasm, which are condu-cive to learning (Cyert & March, 1963; March, 1976).An organization opting to use a stretch goal whenit has greater slack should also experience affec-tive responses that aid in performance improve-ments. Stretch goal pursuit would typically in-volve failures and complications, at least in theearly stages of implementation, and having ade-quate (or more than adequate) resources shouldmake collective commitment more resilient whenthose obstacles surface (Young et al., 1996). A lackof tangible slack resources to credibly attemptstretch goal attainment, however, could easilylower morale and resilience needed to sustaininitiative.

In addition, organizations with greater slackcan test more varied actions and survive fail-ures, which underresourced competitors are lesscapable of doing (e.g., having a capital reserveor back orders on existing products can help anorganization survive a large R&D effort thatfails). Using an organizational stretch goal toimpose a crisis is precarious, since such inten-tionally generated crises meant to spur change“usually overwhelm organizations and theirmembers’ emotional, cognitive and behavioralcapacities” and can result in dysfunctional ri-gidity (Barnett & Pratt, 2000: 75). However, orga-nizations are more likely to avoid rigidity andinstead respond to the imposed crisis with flex-ibility to the extent that they have sufficient timeand encouragement for knowledge generation(Barnett & Pratt, 2000). Having slack resourcesthus permits an organization using stretch goalsto encourage knowledge generation and to morefeasibly and successfully engage in actions thatcan bring about learning from the goal, such assearch that is extensive both in breadth andduration (Sitkin, 1992). Moreover, when excessslack resources are available, parallel initia-tives for stretch goal attainment can be pursuedin different units (Brown & Eisenhardt, 1998;March, 1991; Weick, 1984), allowing for the sys-

tematic and controlled experimentation vital tolearning (Campbell, 1969). When slack is un-available, however, learning is likely to be ham-pered, because a lack of resources constrainsthe capacity to undertake careful trial-and-erroractivities and obtain useful feedback (Cyert &March, 1963; Singh, 1986).

Finally, whereas slack resource availabilityallows organizations to take on new stretch tar-gets without having to abandon what is alreadyworking well, low slack requires organizationspursuing stretch goals to divert resources awayfrom other areas, which if carried out too exten-sively or for too long might eventually result in theloss of beneficial path dependencies and declinein overall performance. Organizations pursuingstretch goals with insufficient slack are also likelyto experience impaired coordination, because co-ordinating actions across units on how to addressradical change requires effort and time, a problemthat is exacerbated as the internal task environ-ment becomes more complex, interdependent,and urgent (Perrow, 1979). Taken together, organi-zations using stretch goals with greater slack re-sources are better situated to experience facilita-tive effects on learning and performance, whereasorganizations without slack are situated for dis-ruptive effects if they use stretch goals.

Proposition 2: For organizations withhigh (low) slack resources, the useof stretch goals will yield positive(negative) effects on learning andperformance.

DETERMINANTS OF STRETCH GOAL PURSUIT

As we argued above, levels of recent perfor-mance and slack resources determine whetheran organization is positioned to experience fa-cilitative or disruptive effects of stretch goals onexploratory learning and performance. But howdo these same organizational factors influencewhether an organization will actually usestretch goals in the first place? In this section weexamine how recent performance and slack rep-resent two distinct motives for pursuing any rad-ical tool or technique: because the organizationmust (e.g., in response to recent performanceproblems) or because the organization can (i.e.,availability of slack resources).


Recent Performance

We argued above that recent organizationalsuccess (performance above a benchmark) sig-nals a greater capacity to reap beneficial out-comes from using stretch goals, but whether or-ganizations with stronger recent performancewill actually be drawn to using such goals is aseparate question. Undertaking a radical toollike a stretch goal is likely to be driven in part byorganizational sensitivity to risk and a propen-sity to accept new challenges, which are af-fected by levels of recent success. Prior researchsuggests that success can breed a conservativedisposition toward risk in organizations (e.g.,Levitt & March, 1988; Sitkin & Pablo, 1992). Thereis an observed tendency for strongly performingorganizations to fall into a success or compe-tency trap, in that success reinforces habits andleads to more exploitation of current skills andpractices and less exploration of new capabili-ties (Lant, Milliken, & Batra, 1992; Levitt & March,1988; Maidique & Zirger, 1985). For example,while consistent success leads to more focuswithin proven routines, Weick (1984), March andOlsen (1976), and Sitkin (1992) argue that it alsoresults in less expansive search activities and areduced willingness to utilize less establishedskills in pursuing riskier, innovative paths. Like-wise, prospect theory (Kahneman & Tversky, 1979),which characterizes the behavior of both individ-uals and organizations (e.g., Mezias, 1988; Singh,1986), implies risk aversion in the domain of gainssuch that when an organization has been perform-ing well, it tends to continue its current strategyand course of action rather than try new or riskyapproaches like stretch goals.

In contrast, an organization experiencing re-cent performance below a benchmark tends tointerpret its present situation as a loss relativeto the benchmark. Under such conditions, theorganization is more likely to undertake riskyactions or even to try radically new methodswith uncertain prospects for success2 (e.g.,

Levinthal & March, 1993; Singh, 1986; Sitkin &Pablo, 1992). Stretch goals may therefore appearmore frequently in organizations that are drivento “go for broke” (Singh, 1986; Sitkin & Pablo,1992: 27) after experiencing noteworthy poor per-formance (called “the failure trap” by Levinthal& March, 1993: 105). Fiegenbaum (1990), Wise-man and Catanach (1997), and Shoham andFiegenbaum (2002), for example, found thatpoorly performing organizations were morelikely to exhibit risky action relative to orga-nizations above the industry average in per-formance. Simon, Houghton, and Savelli (2003)and Miller and Chen (2004) found that organi-zations with disappointing performance un-dertook riskier projects. Lee (1997) found thatorganizations unable to match their own pastperformance exhibited risk seeking relative tothose exceeding their own past performance.All of this suggests that stretch goals would beless likely to be used by successful performersthan by unsuccessful ones.

Proposition 3: Stronger recent perfor-mance is associated with a lower like-lihood that an organization will usestretch goals.

Slack Resources

The ready availability of excess uncommittedresources should position an organization toreap positive effects on learning and perfor-mance should it choose to use stretch goals, aswe argued above. Although slack resources al-low for experimentation with radical tools suchas stretch goals (e.g., Cyert & March, 1963;Nohria & Gulati, 1996), an organization withslack resources is probably unlikely to actuallyuse such goals, because slack functions as aninertia- and complacency-fostering buffer be-tween the organization and the environment(e.g., Sitkin, 1992), rather than as a force for thecreation of an enriching, exploration-enhancingmilieu. Indeed, empirical studies have shownthat the presence of unabsorbed slack inhibitsrisk taking and adaptiveness (e.g., Kraatz & Za-jac, 2001; Maidique & Zirger, 1985).

2 It is important to emphasize that recent performance isthe focus of our argument. Over a longer time horizon, per-formance can cause aspirations to change. Performance thathas been consistently below an industry average or someother reference point can drive down aspirations, resultingin a decreased propensity to undertake risks designed forperformance improvement (Lant, 1992; Mezias, Chen, & Mur-phy, 2002). Similarly, work on threat rigidity (Staw et al., 1981)and permanently failing organizations (Meyer & Zucker,

1989) highlights that some organizations with extreme levelsof poor performance may avoid escalating risky behaviorbecause of an inability to respond (e.g., threat rigidity) or anability to buffer (e.g., permanent failure).


We surmise that “free” resources (i.e., slack)typically are not used by organizations to ex-plore more freely; instead, slack seems to reducethe felt pressure for organizations to be respon-sive. Thus, we expect that having greater slackresults in less motivation to engage in actionsthat involve substantial adaptation or change inroutines. In terms of the behavioral theory of thefirm (Cyert & March, 1963), we posit that organi-zations will not typically engage in expansiveslack-driven search, even in the presence of suf-ficient slack, but, rather, will engage in at mostincremental search and local adaptation, whichwill preclude attraction to the use of radicalinitiatives like stretch goals.

Proposition 4: Greater slack is associ-ated with a lower likelihood that anorganization will use stretch goals.

The Paradox of Stretch Goals

Figure 2 summarizes our propositions con-cerning how different combinations of slack andrecent performance influence the expected va-lence (positive versus negative) of stretch goaleffects on learning and performance (Proposi-tions 1 and 2), as well as the likelihood organi-zations will actually use stretch goals (Proposi-tions 3 and 4). The figure reveals that there is a

misalignment between the organizations wepredict are most likely to reap learning and per-formance benefits from stretch goals and theorganizations most likely to pursue such goals.In cell 2, for example, an organization has boththe slack resources to help take on the risks oftrying a stretch goal and the positive momentumassociated with recent performance success(Propositions 1 and 2). However, when an orga-nization is doing well, we suggest it will gener-ally be averse to aggressive risky or radicalchange initiatives (Propositions 3 and 4). An or-ganization in this cell thus has the lowest rela-tive likelihood of pursuing stretch goals, al-though it is arguably the best situated forfacilitative learning and performance effects. Itis understandable that such a munificent cir-cumstance could obviate the perceived need touse stretch goals, but organizations in such asituation would be advised to fight inertia andoccasionally go against their inclination toavoid stretch goals because they have the req-uisite resources and resilience (in terms of cog-nitive, affective, and behavioral capacity) tobenefit from such goals.

In cell 3, where recent performance and slackare both low, Propositions 3 and 4 would suggestthat stretch initiatives are most likely to be un-dertaken, particularly as a way to reverse past

FIGURE 2The Paradox of Stretch Goals

High

Effect of use: Neutral to disruptive

Likelihood of use: Low

Cell 1

Effect of use: Mostfacilitative

Likelihood of use: Lowest

Cell 2

Slack resources HighLow

Recent

performance

Low

Cell 3 Cell 4

Effect of use: Most disruptive

Likelihood of use: Highest

Effect of use: Neutral tofacilitative

Likelihood of use: High


failures or as a last-ditch effort for survival. De-spite the motivational, performance-driven se-duction of stretch goals when recent perfor-mance and slack are both low, organizations inthis cell are positioned to experience primarilydisruptive effects on learning and performancebecause they do not have the cognitive or affec-tive advantages of recent success (Proposition 1)or the buffer around experimental action thatslack provides (Proposition 2). Counter to whatwe predict will be the inclination of such orga-nizations, they should avoid stretch goals andshould instead pursue “small wins” (Weick,1984) to try to accumulate resources and resil-ience. Or, at most, such organizations could try a“small losses” strategy (Sitkin, 1992) that allowsfor building experience in a way that costs lessand requires less resilience for survival.

Figure 2 also reveals that for two combina-tions of slack and recent performance (cells 1and 4), the likelihood of using stretch goals isreasonably aligned with the expected positiveor negative valence of effects. In cell 1, organi-zations with strong recent performance and lowslack should have a relatively low propensity topursue stretch goals (as compared to cells 3 or 4).Recalling Proposition 3, organizations with verystrong recent performance may become mired incompetency traps and tend to avoid riskychange. These organizations simply do not havea pressing need for stretch goals in the nearterm. Although Proposition 4 suggests that lessslack tends to be associated with a higher pro-pensity to pursue stretch goals, we expect thatthis tendency is tempered significantly by per-formance-driven conservatism. Because of thepowerful effects of recency on judgment (e.g.,Bjork & Whitten, 1974; Hogarth & Einhorn, 1992),we surmise that recent performance will be astronger factor than slack when it comes to whatdrives an organization to act, and, thus, organi-zations with strong recent performance will ex-hibit a relatively low likelihood to pursuestretch goals.3 Organizations in this cell should,in fact, follow their strategic inclination to avoidusing stretch goals. Although strong recent per-formance would provide some advantages inleveraging the positive potential of stretch goals

(Proposition 1), limited unabsorbed slack re-sources act as a significant constraint thatwould contribute to primarily disruptive (or atbest neutral) effects of stretch pursuit on learn-ing and performance (Proposition 2).

In cell 4, where the level of slack resources ishigh but recent performance is weak, organiza-tions will exhibit a relatively high likelihood ofpursuing stretch goals (compared to cells 1 or 2).For organizations in this cell, the pursuit ofstretch initiatives is driven by the desire to re-coup recent performance losses (Proposition 3),without the tempering effect of (slack) resourceconstraints. Although higher levels of slackcould contribute to complacency and dampenenthusiasm for radical change (Proposition 4),we once again expect that the power of re-cency (e.g., Bjork & Whitten, 1974; Hogarth &Einhorn, 1992) is likely to motivate organiza-tional decision makers to emphasize recentpoor performance, even over successful (butfar less salient) longer-term past performance.Organizations in this cell also have some ad-vantages that should protect them from disrup-tive stretch goal effects on learning and perfor-mance. Greater slack resources provide buffersagainst short-run performance problems andmake the organization more resilient in the faceof the extreme demands of stretch goals, and theongoing munificence of the internal environ-ment makes large-scale, risky projects more fea-sible to undertake (Proposition 2). Overall, whenreacting to recent performance losses and ade-quately resourced with slack, organizationsare reasonably likely to use stretch goals andare also positioned to experience some facilita-tive (or at least neutral) effects on learning andperformance.

Our analysis reveals a paradox: the organiza-tions most likely to benefit from stretch goals areleast likely to use them, and the organizationsleast likely to benefit from them are the most likelyto use them (i.e., cells 2 and 3 in Figure 2). Thispattern suggests that it is really only under quitelimited conditions that organizations will besafely positioned to experience positive learningand performance outcomes from pursuing stretchtargets—namely, organizations with both highslack resources and high recent performance (cell2 in Figure 2). Yet few organizations can realisti-cally be expected to fall into that category. In fact,we speculate that most organizations carry lowlevels of unabsorbed slack (cells 1 or 3), which by

3 Recency effects (e.g., Bjork & Whitten, 1974; Hogarth &Einhorn, 1992) in psychology refer to a cognitive tendencywhere recent stimuli or events are more salient and, thus, aregiven disproportionate weight in judgment and decisions.


our analysis would suggest that stretch goals usu-ally will have disruptive, or at best neutral, effectson learning and performance for the typicalorganization.

DISCUSSION

Drawing from a variety of theoretical perspec-tives, our primary purpose in this article is to ad-vance the understanding of exploratory learningand adaptation by systematically investigatingthe causes and effects of pursuing seemingly im-possible organizational goals—stretch goals. Inthis section we review our conclusions and exam-ine the contributions our work makes to extanttheory and research in several areas of inquiry.We also consider the reasons behind the commonbelief that stretch goals are simple and widelysuccessful when careful analysis suggests a morecomplicated picture. Finally, we identify researchopportunities that extend beyond the scope of thepresent investigation and highlight implicationsfor managerial practice.

Opening the Theoretical Black Box ofSeemingly Impossible Goals

Goals and aspirations play an important rolein organizational theory (e.g., Cyert & March,1963). Yet scholars have placed little emphasison understanding the specific underlying pro-cesses through which macrolevel goals, partic-ularly unattainable goals, can influence organi-zational outcomes. Thus, our theoreticalinvestigation of how stretch goals function andaffect organizational outcomes began with athorough consideration of the cognitive, affec-tive, and behavioral mechanisms by whichseemingly impossible goals can influence ex-ploratory learning and organizational perfor-mance (summarized in Figure 1). Going beyondprior thinking on organizational stretch goals,which has not systematically considered mech-anisms and has focused largely on presumedpositive effects, we examined the potential forseemingly impossible goals to both facilitateand disrupt learning and performance. We thenproposed that whether facilitative or disruptiveeffects ensue is contingent on two core organi-zational factors: recent performance and slackresources (Propositions 1 and 2). We also positedthat these same two factors determine the like-lihood that organizations will be drawn to usingstretch goals (Propositions 3 and 4).

The conclusion suggested by the aggregatedanalysis of our four propositions, as revealed inFigure 2, is the unfortunate tendency for thewrong organizations to be most drawn to usingstretch goals. Furthermore, whereas weak organi-zations might pursue stretch goals out of desper-ation and make their dire circumstances worse,those with the capabilities to truly benefit fromstretch goals typically fail to do so because thesame characteristics that make them well posi-tioned to benefit from stretch goals also limit theirinclination to actually pursue them. This situationis what we call “the paradox of stretch goals.”

Our theory building contributes interdisci-plinary insights at the intersection of the litera-ture on organizational goals, on learning, andon adaptation by investigating the means bywhich extreme goals can prompt exploration(e.g., Rahmandad, 2008; Uotila et al., 2009) anddynamic capability development (e.g., Agarwal& Helfat, 2009; Capron & Mitchell, 2009; Winter,2000). As summarized in Figure 1, our analysisindicates that stretch goals can capture atten-tion and facilitate openness to questioning thevalidity of old assumptions, old information,and old frameworks. Stretch goals can alsoprompt behaviors traditionally associated withlearning, such as broad search, experimenta-tion, and trial and error. Yet, at the same time,some of the most critical conditions for effectivelearning to occur (such as more experience, con-trollable change, and interpretable feedback) donot automatically hold in the case of stretchgoals, and, thus, organizations are unlikely tolearn from them unless they have the slack re-sources to repeatedly and systematically exper-iment and withstand failures.

The requisite resilience required to effectivelypursue stretch goals is also aided by strong recentperformance. These ideas provide a basis fordeeper analysis and clarity around the processesand conditions needed for organizations to learneffectively under ambiguity (March & Olsen, 1976;Weick, 1979, 1995). Although we build on the be-havioral theory of the firm (Cyert & March, 1963) insuggesting that slack resources can positively en-able organizations to benefit from using stretchgoals (Proposition 2), our predictions diverge fromthe behavioral theory of the firm when we positthat organizations with these capabilities are un-likely to actually use innovation methods that areas radical and risky as stretch goals (Proposition4). That is, we diverge because we expect that


having greater unabsorbed slack, while enabling,results in less motivation to actually engage inactions that involve substantial adaptation, risk,or change in routines.

In considering the reasons why organizationsmight be drawn to use stretch goals, our analy-sis speaks to the literature on organizationalchange and adaptation to internal and externaljolts (e.g., Barnett & Pratt, 2000; Beer et al., 1990;D’Aveni & MacMillan, 1990; Meyer, 1982). We ar-gue that stretch goals can be viewed as auto-genic crises meant to spur innovation, change,learning, and increases in performance (e.g.,Barnett & Pratt, 2000; D’Aveni & MacMillan,1990). Our theoretical analysis of the causes andeffects of using seemingly impossible goalsgenerates additional insight into how organiza-tions more or less effectively learn from bothinternal and environmental stimuli (e.g., Sut-cliffe, 1994), as well as the conditions that giverise to punctuated organizational change pro-cesses (Romanelli & Tushman, 1994).

Finally, our investigation advances our under-standing of the effects of unattainable goals(e.g., Garland, 1982, 1983; Locke, 1982, 2004; Rous-seau, 1997) at a collective level by systemati-cally surfacing specific mechanisms throughwhich such goals have organizational out-comes. As such, our analysis contributes to theliterature on organizational goals (e.g., Cyert &March, 1963; Ethiraj & Levinthal, 2009; Lant, 1992;Lant & Shapira, 2008) by revealing how stretchgoals can challenge important scholarly as-sumptions regarding the relationship betweenaspirations and firm performance. We arguethat stretch goals create aspiration-performancegaps that are too extreme to reliably lead toorganizational performance increases, exceptfor the rare organizations that have sufficientslack resources and strong recent performance.

The Persistent Perception That Stretch GoalsConsistently Lead to Success

Because the scholarly research we draw fromsuggests that the effects of seemingly impossi-ble goals are complex and contingent, we needto ask why so many authors (mostly, but notexclusively, practice-focused authors) arguethat stretch goals are an unmitigated organiza-tional success story. Numerous practitioner-focused writings assert that stretch goals reli-ably lead to positive performance results (e.g.,

Collins & Porras, 1994; Hamel, 1998; Hamel &Prahalad, 1993), including reports concerning3M, CSX, Motorola, General Electric, Union Pa-cific, Boeing, Mead, and Toyota (Takeuchi et al.,2008; Thompson et al., 1997; Tully, 1994). Suchhighly visible successes make salient the per-ception of universal stretch goal success, as il-lustrated in the following effusive endorsementfrom Jack Welch, former CEO of General Electric(where the term stretch may have originated):

We have what we call stretch targets. . . . For ex-ample, we spent 105 years in this company andwe never had double digit operating margins. Wesaid in 1991 we want 15 and we put it in ourannual report. We told everybody. 9.6 or 9.5 wasour best. We’ll do 14 in 1994, and prices have beengoing down in the global market. And we’ll do 15next year. We never did more than 5 inventoryturns and we said we’ll do 10. We had no ideahow we would get to 10. . . . The big line I usetoday is that budgets enervate and stretch ener-gizes. It’s real (Bartlett, 1999).

There are several reasons why stretch goals maybe persistently perceived and portrayed as morebeneficial for learning and performance than wepropose they actually are. Our theoretical analysissuggests one particular reason why this mispercep-tion may arise and persist: most famous stretch goalsuccess stories come from organizations in whichmanagement set very ambitious performance goalsto prepare for a radical change while the organiza-tion had good resource endowments (e.g., IBM in the1960s, General Electric in the 1980s, and Toyota inthe 1990s; see also Bartlett & Wozny, 1999; Hamel,1998; Sherman, 1995; Takeuchi et al., 2008; Thompsonet al., 1997). That is, proponents of stretch goals mayhave overgeneralized based on evidence from orga-nizations that had substantial slack resources and,in many cases, also had strong recent performance(i.e., cell 2 in Figure 2). By our analysis, these orga-nizations were unusually well positioned to benefitfrom stretch goals (Propositions 1 and 2). But theseorganizations were probably outliers that somehowavoided the dominant pattern of inertia that Propo-sitions 3 and 4 would predict for organizations hav-ing both a large amount of slack and strong recentperformance. Thus, we argue that the generalizedbasis for the success of stretch goals has been mis-construed and has consequently led to an oversim-plified and inflated perception of their value.

Aside from focusing on high-performing organiza-tions with substantial resources, the pervasive viewof stretch goals as beneficial could be due to successstories based on incomplete learning cycles, such as


when apparent consequences are actually unre-lated to organizational action and lead to whatMarch and Olsen (1976) call “superstitious” experi-ential learning (see also Huber, 1991). Similarly,some organizational decision makers may come tounjustified or overconfident conclusions based onfew data points, a sample size problem where faultylearning can occur from low-frequency events inwhich performance effects are often confoundedwith random error or other fallacious data (March,Sproull, & Tamuz, 1991). Moreover, failure myopia(Levinthal & March, 1993) may be an issue, wherebyorganizational learning is based on oversampling ofsuccesses and undersampling of failures becausesuccessful practitioner examples, salient and com-pelling as they are, may be but a small, unrepresen-tative drop in a large ocean of organizational initia-tives (see also Denrell, 2003). Base-rate neglect canalso make highly risky exploration (e.g., stretchgoals) appear to be a good idea, with an unrealisti-cally high expectation of success. In addition, orga-nizations that pursue stretch goals, particularlythose that do so as a last-ditch effort when they arealready failing, might subsequently fail entirely anddisappear, thus biasing the observable sample ofusers toward success. Observers trying to judge theeffectiveness of stretch goals are therefore exposedto a systematically biased sample with an overrep-resentation of successes.

All of these issues lead us to conclude that inves-tigations based on a few select cases, includingmany that appear in practice-focused publications,are unlikely to present a balanced picture of stretchgoal processes and effects. By sidestepping glib ex-hortations and focusing instead on a generative,theoretically grounded analysis of mechanisms andcontingency factors, we hope we have providedsome useful guidelines for managers to use in tai-loring their practices to conditions most likely tolead to positive outcomes.

Foundations for Future Research

We conclude with a discussion of possible em-pirical settings and extensions for research onstretch goals, as well as a brief commentary ontwo implications of stretch goal pursuit that arebeyond the scope of the current investigation buthighlight important areas for future inquiry.

Empirical tests of stretch goal effects. To setthe stage for future empirical work, we haveproposed a testable theory suggesting thatseemingly impossible organizational goals ex-

hibit a predictable pattern of positive and neg-ative effects contingent on recent performanceand slack resources, and these contingency fac-tors also determine when organizations aremore or less likely to use such goals. Our prop-ositions could be examined using several em-pirical approaches. For example, initial empiri-cal studies of stretch goal effects could take theform of qualitative theory testing, whereby mul-tiple firms with different attributes are studiedin depth. Or a study using one very large orga-nization could examine the use and effective-ness of project proposals within the organiza-tion (e.g., high-risk research initiatives or newventure proposals). Alternatively, a simulationstudy would provide the control to vary organi-zational conditions and measure baseline levelsof learning and performance before and afterthe imposition of a stretch goal.

In addition, quantitative investigations couldbe undertaken to study both the effects and like-lihood of pursuing stretch goals using a strati-fied random sample based on archival indica-tors of slack and recent performance. Within thesample, evidence of stretch goal usage could bemeasured by conducting a survey of senior man-agers in organizations, or it could be extractedfrom annual reports or letters to shareholders. Inany of these approaches, individual organiza-tions in the sample could be coded for the timingof stretch goal use and the type or nature of thestretch goal (e.g., a goal to increase marketshare, reduce manufacturing process time, re-duce error rate) in order to better compare ef-fects across organizations.

Beyond testing the organization-level effects wehave proposed in this article, future researchcould focus more specifically on the potential un-derlying processes and mechanisms summarizedin Figure 1. That is, researchers could study unat-tainable goals in terms of their impact on thecognitive, affective, and behavioral mechanismsthrough which stretch goals can influence organi-zational outcomes. To further enrich and extendthe theory proposed here, future studies could alsoexamine possible sequences, interactions, andpath dependencies among the mechanisms. Itwould be interesting as well to explore the impli-cations of variation in the degree of stretch goalimplementation, such as when unattainable goalsare used only by segments of an organization ver-sus the entire organization.


In our theorizing we have discussed organiza-tion-level effects on learning and performance,but within the organization stretch goals mighthave differential effects on various parts of thesystem. Additionally, because the use of stretchgoals implies a willingness to undertake highrisks to pursue high rewards, variation in incen-tives within organizations for the pursuit of stretchgoals could be examined to see how this affectsboth use and effectiveness of such goals. All ofthese questions highlight fruitful ways in whichthe ideas we have proposed can be further ex-tended and refined if future research were to focuson the ripple effects inside organizations that usestretch goals, rather than focusing only on theaggregated organization-level outcomes thathave been our primary focus.

Process-oriented stretch goals. Scholarly andpractitioner accounts of stretch goals suggestthat such goals are typically used in the explicithope of directly improving performance, and po-tentially also spurring exploratory learning inthe course of achieving the extreme target. Con-sistent with the use of stretch goals in practiceand also the majority of goal research, we havedefined and analyzed stretch goals as outcome-oriented goals that make explicit a specific,quantifiable level of desired performance or out-put (e.g., decrease errors by a particular percent-age). Note that this precludes goals that arearticulated solely in terms of enhancing thelearning process as an end in itself (e.g., wherethe goal is to formulate five ways of learninghow to decrease errors). This raises the questionof whether learning-focused organizations (i.e.,considering the use of stretch goals solely tolearn) really need to use stretch goals (which areperformance outcome goals) as a vehicle, orwhether they could gain the facilitative benefitsof stretch goals and limit their disruptive liabil-ities by using process-oriented learning goalsinstead (see, for example, Seijts & Latham, 2005).It is ultimately an empirical question, and futurework could explore whether our analysis holdswhen one moves beyond outcome-oriented per-formance goals to include seemingly impossiblegoals that are explicitly focused on the acquisi-tion of procedural knowledge at the organiza-tion level of analysis.

Ethical implications of stretch goal pursuit.While we have focused on the organization levelof analysis, we wish to point out some findingsat the individual level pertaining to the unin-

tended negative consequences of goal setting,with a particular focus on how goals can induceunethical behavior (Barsky, 2008; Jensen, 2003;Locke, 2004; Ordonez, Schweitzer, Galinsky, &Bazerman, 2009; Schweitzer, Ordonez, & Douma,2004). Although goals can pose an especiallydangerous temptation to breach ethics when as-sociated with performance measurement orcompensation systems by creating incentives tomisrepresent performance (Jensen, 2003; Locke,2004; Schweitzer et al., 2004), empirical work pro-vides direct evidence that even nonmonetaryrewards for meeting a goal (such as internalsatisfaction) are sufficient to influence unethicalbehavior (Schweitzer et al., 2004).

Moving from merely difficult goals at the in-dividual level to the case of stretch goals at theorganization level, the ethical implicationscould be amplified. The seeming impossibilityof achieving stretch goals using ordinary meanscould easily slip into the seductiveness of usingextraordinary means, such as falsifying records.Managing the unintended ethical perils ofseemingly impossible goals, as well as otherindividual-level disruptive effects that are be-yond the scope of the present investigation, isworthy of further examination.

CONCLUSION

We investigated the organizational pursuit ofseemingly impossible goals—commonly knownas stretch goals—and the effects of such goalson exploratory learning and increased perfor-mance. Despite widespread interest in the ideaof stretch goals and assertions about their sys-tematic benefits, there are many reasons to in-terpret the normative stories of stretch goal suc-cess as resulting from misunderstandings of theactual functioning and effects of this enticingtechnique. Although our analysis suggests thatthere are very specific and limited conditionsunder which some organizations should pursueseemingly impossible goals, we recognize thatorganizations may continue to look for successin all the wrong places. Thus, a challenge fororganizational researchers is to more clearlyidentify control mechanisms, learning strate-gies, and decision aids that organizations canuse to achieve greater effectiveness in overcom-ing the paradox of stretch goals.


REFERENCES

Adler, P. S., & Obstfeld, D. 2007. The role of affect in creativeprojects and exploratory search. Industrial and Corpo-rate Change, 16: 19–50.

Agarwal, R., & Helfat, C. E. 2009. Strategic renewal of orga-nizations. Organization Science, 20: 281–293.

Argote, L. 1999. Organizational learning: Creating, retaining,and transferring knowledge. New York: Kluwer.

Argyris, C. 1985. Strategy, change, and defensive routines.Boston: Pitman.

Argyris, C., & Schon, D. A. 1978. Organizational learning: Atheory of action perspective. Reading, MA: Addison-Wesley.

Atkinson, J. W. 1964. An introduction to motivation. Princeton,NJ: Van Nostrand.

Audia, P. G., Locke, E. A., & Smith, K. G. 2000. The paradox ofsuccess: An archival and a laboratory study of strategicpersistence following radical environmental change.Academy of Management Journal, 43: 837–851.

Barnett, C. K., & Pratt, M. G. 2000. From threat-rigidity toflexibility—Toward a learning model of autogenic crisisin organizations. Journal of Organizational ChangeManagement, 13: 74–88.

Barsade, S. G. 2002. The ripple effect: Emotional contagionand its influence on group behavior. Administrative Sci-ence Quarterly, 47: 644–675.

Barsade, S. G., & Gibson, D. E. 2007. Why does affect matterin organizations? Academy of Management Perspec-tives, 21(1): 36–59.

Barsky, A. 2008. Understanding the ethical cost of organiza-tional goal-setting: A review and theory development.Journal of Business Ethics, 81: 63–81.

Bartel, C. A., & Saavedra, R. 2000. The collective construction ofwork group moods. Administrative Science Quarterly, 45:197–231.

Bartlett, C. A. 1999. GE compilation: Jack Welch 1981–1999.Video, Harvard Business School, Boston.

Bartlett, C. A., & Wozny, M. 1999. GE’s two-decade transfor-mation: Jack Welch’s leadership. Case No. 301040-MMC-ENG. Boston: Harvard Business School Case Services.

Bass, B., & Riggio, E. G. 2005. Transformational leadership(2nd ed.). Mahwah, NJ: Lawrence Erlbaum Associates.

Baum, J. A. C., & Dahlin, K. B. 2007. Aspiration performanceand railroads’ pattern of learning from train wrecks andcrashes. Organization Science, 18: 368–385.

Baumard, P., & Starbuck, W. H. 2005. Learning from failures:Why it may not happen, Long Range Planning, 38: 281–298.

Beer, M., Eisenstat, R. A., & Spector, B. 1990. The critical pathto corporate renewal. Boston: Harvard Business SchoolPress.

Bjork, R. A., & Whitten, W. B. 1974. Recency-sensitive retrievalprocesses in long-term free recall. Cognitive Psychol-ogy, 6: 173–189.

Bourgeois, L. J. 1981. On the measurement of organizationalslack. Academy of Management Review, 6: 29–39.

Brown, S. L., & Eisenhardt, K. M. 1998. Competing on the edge:Strategy as structured chaos. Boston: Harvard BusinessSchool Press.

Campbell, D. T. 1969. Reforms as experiments. AmericanPsychologist, 24: 409–429.

Capron, L., & Mitchell, W. 2009. Selection capability: Howcapability gaps and internal social frictions affect inter-nal and external strategic renewal. Organization Sci-ence, 20: 294–312.

Carley, K. M., & Lin, Z. A. 1997. A theoretical study of orga-nizational performance under information distortion.Management Science, 43: 976–997.

Chesney, A. A., & Locke, E. A. 1991. Relationships amonggoal difficulty, business strategies, and performance ona complex management simulation task. Academy ofManagement Journal, 34: 400–424.

Christensen, C. M. 1997. The innovator’s dilemma: When newtechnologies cause great firms to fail. Boston: HarvardBusiness School Press.

Cohen, W. M., & Levinthal, D. A. 1990. Absorptive capacity: Anew perspective on learning and innovation. Adminis-trative Science Quarterly, 35: 128–152.

Colbert, A. E., Kristof-Brown, A. I., Bradley, B. H., & Barrick,M. R. 2008. CEO transformational leadership: The role ofgoal importance congruence in top management teams.Academy of Management Journal, 51: 81–96.

Collins, J. C., & Porras, J. I. 1994. Built to last: Successfulhabits of visionary companies. New York: HarperCollins.

Collins, J. C., & Porras, J. I. 1996. Building your company’svision. Harvard Business Review, 74(5): 65–77.

Cyert, R. M., & March, J. G. 1963. Behavioral theory of the firm.Englewood Cliffs, NJ: Prentice-Hall.

D’Aveni, R. A., & MacMillan, I. C. 1990. Crisis and the contentof managerial communications: A study of the focus ofattention of top managers in surviving and failing firms.Administrative Science Quarterly, 35: 634–657.

Denrell, J. 2003. Vicarious learning, undersampling of failure,and the myths of management. Organization Science,14: 227–243.

Dess, G. G. 1987. Consensus on strategy formulation and orga-nizational performance: Competitors in a fragmented in-dustry. Strategic Management Journal, 8: 259–277.

Drazin, R., & Sandelands, L. 1992. Autogenesis: A perspectiveon the process of organizations. Organization Science, 3:230–249

Dvir, T., Eden, D., Avolio, B. J., & Shamir, B. 2002. Impact oftransformational leadership on follower developmentand performance: A field experiment. Academy of Man-agement Journal, 45: 735–744.

Edmondson, A. C., Bohmer, R. M., & Pisano, G. P. 2001. Dis-puted routines: Team learning and new technology im-plementation in hospitals. Administrative Science Quar-terly, 46: 685–716.

Ethiraj, S. K., & Levinthal, D. 2009. Hoping for A to Z while


rewarding only A: Complex and multiple organizationalgoals. Organization Science, 20: 4–21.

Fang, C., & Levinthal, D. 2009. The near term liability ofexploitation: Exploration and exploitation in multi-stageproblems. Organization Science, 20: 538–551.

Ferrier, W. J. 2001. Navigating the competitive landscape:The drivers and consequences of competitive aggres-siveness. Academy of Management Journal, 44: 858–877.

Fiegenbaum, A. 1990. Prospect theory and the risk-returnassociation: An empirical examination in 85 indus-tries. Journal of Economic Behavior and Organization,14: 187–203.

Freiberg, K., & Freiberg, J. 1996. Nuts! Southwest Airlines’crazy recipe for business and personal success. Austin,TX: Bard Press.

Galbraith, J. R. 1973. Designing complex organizations. Read-ing, MA: Addison-Wesley.

Garland, H. 1982. Goal levels and task performance: A com-pelling replication of some compelling results. Journalof Applied Psychology, 67: 245–248.

Garland, H. 1983. Influence of ability, assigned goals, andnormative information on personal goals and perfor-mance: A challenge to the goal attainability assump-tion. Journal of Applied Psychology, 68: 20–30.

Golovin, J. J. 1997. Achieving stretch goals: Best practices inmanufacturing for the new millennium. Upper SaddleRiver, NJ: Prentice-Hall.

Greve, H. R. 1998. Performance, aspirations, and risky organi-zational change. Administrative Science Quarterly, 43: 58–77.

Greve, H. R. 2003. Organizational learning from performancefeedback. Cambridge: Cambridge University Press.

Hamel, G. 1998. Strategy innovation and the quest for value.Sloan Management Review, 39(2): 7–15.

Hamel, G., & Prahalad, C. K. 1993. Strategy as stretch andleverage. Harvard Business Review, 71(2): 75–84.

Heath, C., Larrick, R. P., & Wu, G. 1999. Goals as referencepoints. Cognitive Psychology, 38: 79–109.

Henderson, R. M., & Clark, K. B. 1990. Architectural innova-tion: The reconfiguration of existing product technolo-gies and the failure of established firms. AdministrativeScience Quarterly, 35: 9–30.

Hitt, M. A. 1988. The measuring of organizational effective-ness: Multiple domains and constituencies. Manage-ment International Review, 28: 28–40.

Hogarth, R. M., & Einhorn, H. J. 1992. Order effects in beliefupdating: The belief-adjustment model. Cognitive Psy-chology, 24: 1–55.

Huber, G. P. 1991. Organizational learning: The contributingprocesses and the literatures. Organization Science, 2: 88–115.

Hughes, R. E. 2001. Contingent use of stretch goals: Consid-erations of workflow integration and risk deviation.Work Study, 50: 7–12.

Ingram, P., & Baum, J. A. C. 1997. Opportunity and constraint:Organizations’ learning from the operating and compet-

itive experience of industries. Strategic ManagementJournal, 18(Summer Special Issue): 75–98.

Janis, I. L., & Mann, L. 1977. Decision making. New York: FreePress.

Jensen, M. C. 2003. Paying people to lie: The truth about thebudgeting process. European Financial Management, 9:379–406.

Kahneman, D., & Tversky, A. 1979. Prospect theory: An anal-ysis of decision under risk. Econometrica, 47: 263–291.

Kang, S. C., Morris, S. S., & Snell, S. A. 2007. Relationalarchetypes, organizational learning, and value creation.Academy of Management Review, 32: 236–256.

Kerr, S., & Landauer, S. 2004. Using stretch goals to promoteorganizational effectiveness and personal growth: Gen-eral Electric and Goldman Sachs. Academy of Manage-ment Executive, 18(4): 134–138.

Kim, L. 1998. Crisis construction and organizational learning:Capability building in catching-up at Hyundai Motor.Organization Science, 9: 506–521.

Kim, T., & Rhee, M. 2009. Exploration and exploitation: Inter-nal variety and environmental dynamism. Strategic Or-ganization, 7: 11–41.

Kotter, J. P. 2008. A sense of urgency. Boston: Harvard Busi-ness School Press.

Kraatz, M. S., & Zajac, E. J. 2001. How organizational re-sources affect strategic change and performance in tur-bulent environments: Theory and evidence. Organiza-tion Science, 12: 632–657.

Lant, T. K. 1992. Aspiration level adaptation: An empiricalexploration. Management Science, 38: 623–644.

Lant, T. K., Milliken, F. J., & Batra, B. 1992. The role of mana-gerial learning and interpretation in strategic persis-tence and reorientation: An empirical exploration. Stra-tegic Management Journal, 13: 585–608.

Lant, T. K., & Shapira, Z. 2008. Managerial reasoning aboutaspirations and expectations. Journal of Economic Be-havior and Organization, 66: 60–73.

Lawrence, P. R., & Lorsch, J. W. 1967. Organization and envi-ronment: Managing differentiation and integration. Bos-ton: Harvard Business School Press.

Lee, D. Y. 1997. The impact of poor performance on risk-taking attitudes: A longitudinal study with PLS causalmodeling approach, Decision Sciences, 28: 59–80.

Leonard-Barton, D. 1992. Core capabilities and core rigidi-ties: A paradox in managing new product development.Strategic Management Journal, 13: 111–125.

Levinthal, D. A., & March, J. G. 1981. A model of adaptiveorganizational search. Journal of Economic Behaviorand Organization, 2: 307–333.

Levinthal, D. A., & March, J. G. 1993. The myopia of learning.Strategic Management Journal, 14: 95–112.

Levitt, B., & March, J. G. 1988. Organizational learning. An-nual Review of Sociology, 14: 319–340.

Locke, E. A. 1982. Relation of goal level to performance witha short work period and multiple goal levels. Journal ofApplied Psychology, 67: 512–514.


Locke, E. A. 2004. Linking goals to monetary incentives.Academy of Management Executive, 18(4): 130–133.

Locke, E. A., & Latham, G. P. 1990. A theory of goal setting andtask performance. Englewood Cliffs, NJ: Prentice-Hall.

Locke, E. A., & Latham, G. P. 1994. Goal setting theory. InH. F. O’Neil & M. Drillings (Eds.), Motivation: Theoryand research: 13–29. Hillsdale, NJ: Lawrence ErlbaumAssociates.

Maidique, M. A., & Zirger, B. J. 1985. The new product learn-ing cycle. Research Policy, 14: 299–313.

March, J. G. 1976. The technology of foolishness. In J. G. March& J. P. Olsen (Eds.), Ambiguity and choice in organizations:69–81. Bergen, Norway: Universitetsforlaget.

March, J. G. 1991. Exploration and exploitation in organiza-tional learning. Organization Science, 2: 71–87.

March, J. G., & Olsen, J. P. 1976. Organizational learning andthe ambiguity of the past. In J. G. March & J. P. Olsen(Eds.), Ambiguity and choice in organizations: 54–68.Bergen: Universitetsforlaget.

March, J. G., Sproull, L. S., & Tamuz, M. 1991. Learning fromsamples of one or fewer. Organization Science, 2: 1–13.

Martin, J. 1992. Organizational cultures. London: Sage.

McGrath, R.G. 2001. Exploratory learning, innovative capac-ity, and managerial oversight. Academy of ManagementJournal, 44: 118–131.

Meyer, A. D. 1982. Adapting to environmental jolt. Adminis-trative Science Quarterly, 27: 515–537.

Meyer, M. W., & Zucker, L. G. 1989. Permanently failing or-ganizations. Newbury Park, CA: Sage.

Mezias, S. J. 1988. Aspiration level effects: An empirical in-vestigation. Journal of Economic Behavior and Organi-zation, 10: 389–400.

Mezias, S. J., Chen, Y. R., & Murphy, P. R. 2002. Aspirationlevel adaptation in an American financial services or-ganization: A field study. Management Science, 48:1285–1300.

Miller, C. C., & Cardinal, L. B. 1994. Strategic planning andfirm performance: A synthesis of more than two de-cades of research. Academy of Management Journal,37: 1649 –1665.

Miller, K. D., & Chen, W. 2004. Variable organizational riskpreferences: Test of the March-Shapira model. Academyof Management Journal, 47: 105–115.

Nanus, B. 1992. Visionary leadership: Creating a compellingsense of direction for your organization. San Francisco:Jossey-Bass.

Natividad, G. 2009. Financial slack, organization, and com-petition in movie distribution. Working paper, New YorkUniversity, New York.

Nohria, N., & Gulati, R. 1996. Is slack good or bad for inno-vation? Academy of Management Journal, 39: 1245–1264.

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

Ordonez, L. D., Schweitzer, M. E., Galinsky, A. D., & Bazer-man, M. H. 2009. Goals gone wild: The systematic side

effects of overprescribing goal setting. Academy of Man-agement Perspectives, 23(1): 6–16.

O’Reilly, C. A. 1982. Variations in decision makers’ use ofinformation sources: The impact of quality and accessi-bility of information. Academy of Management Journal,25: 756–771.

Perrow, C. 1979. Complex organizations. Glenview, IL: ScottForesman.

Quinn, R. E., & Rohrbaugh, J. 1983. A spatial model of effective-ness criteria: Towards a competing values approach toorganizational analysis. Management Science, 29: 363–377.

Rahmandad, H. 2008. Effect of delays on complexity of orga-nizational learning. Management Science, 52: 1297–1312.

Raisch, S., & Birkinshaw, J. 2008. Organizational ambidexter-ity: Antecedents, outcomes, and moderators. Journal ofManagement, 34: 375–409.

Romanelli, E., & Tushman, M. 1994. Organizational transfor-mation as punctuated equilibrium: An empirical test.Academy of Management Journal, 37: 1141–1166.

Rousseau, D. M. 1997. Organizational behavior in the neworganizational era. Annual Review of Psychology, 48:515–546.

Schweitzer, M., Ordonez, L., & Douma, B. 2004. Goal setting asa motivator of unethical behavior. Academy of Manage-ment Journal, 47: 422–432.

Seijts, G. H., & Latham, G. P. 2005. Learning versus perfor-mance goals: When should each be used? Academy ofManagement Executive, 19(1): 124–131.

Shamir, B., House, R. J., & Arthur, M. B. 1993. The motivationaleffects of charismatic leadership: A self-concept basedtheory. Organization Science, 4: 577–594.

Sherman, S. 1995. Stretch goals: The dark side of asking formiracles. Fortune, November 13: 231–232.

Shoham, A., & Fiegenbaum, A. 2002. Competitive determinantsof organizational risk-taking attitude: The role of strategicreference points. Management Decision, 40(2): 127–141.

Siggelkow, N., & Levinthal, D. A. 2003. Temporarily divide toconquer: Centralized, decentralized, and reintegratedorganizational approaches to exploration and adapta-tion. Organization Science, 14: 650–669.

Simon, M., Houghton, S. M., & Savelli, 2003. Out of the fryingpan . . .? Why small business managers introduce high-risk products. Journal of Business Venturing, 18: 419–440.

Singh, J. V. 1986. Performance, slack, and risk taking in or-ganizational decision making. Academy of Manage-ment Journal, 29: 562–585.

Sitkin, S. B. 1992. Learning through failure: The strategy ofsmall losses. Research in Organizational Behavior, 14:231–266.

Sitkin, S. B., & Pablo, A. L. 1992. Reconceptualizing the deter-minants of risk behavior. Academy of Management Re-view, 17: 9–38.

Sitkin, S. B., Sutcliffe, K. M., & Schroeder, R. G. 1994. Distin-guishing control from learning in Total Quality Manage-ment: A contingency perspective. Academy of Manage-ment Review, 18: 537–564.


Sitkin, S. B., Sutcliffe, K. M., & Weick, K. E. 1998. Organiza-tional learning. In R. C. Dorf (Ed.), The technology man-agement handbook: 70–76. Boca Raton, FL: CRC Pressand IEEE Press.

Smith, K. G., Locke, E. A., & Barry, D. 1990. Goal setting,planning, and organizational performance: An experi-mental simulation. Organizational Behavior and Hu-man Decision Processes, 46: 118–135.

Starbuck, W. H. 1983. Organizations as action generators.American Sociological Review, 48: 91–102.

Staw, B. M., Sandelands, L. E., & Dutton, J. E. 1981. Threat-rigidity effects in organizational behavior: A multi-level analysis. Administrative Science Quarterly, 26:501–524.

Sutcliffe, K. M. 1994. What executives notice: Accurate per-ceptions in top management teams. Academy of Man-agement Journal, 37: 1360–1378.

Sutcliffe, K. M., Sitkin, S. B., & Browning, L. D. 2000. Tailoringprocess management to situational requirements: Be-yond the control and exploration dichotomy. In R. Cole &W. R. Scott (Eds.), The quality movement and organiza-tion theory: 315–330. Thousand Oaks, CA: Sage.

Takeuchi, H., Osono, E., & Shimizu, N. 2008. The contradic-tions that drive Toyota’s success. Harvard Business Re-view, 86(6): 96–104.

Thompson, K. R., Hochwarter, W. A., & Mathys, N. J. 1997.Stretch targets: What makes them effective? Academy ofManagement Executive, 11(3): 48–60.

Tully, S. 1994. Why go for stretch targets? Fortune, November14: 145–158.

Uotila, J., Maula, M., Keil, T., & Zahra, S. A. 2009. Exploration,exploitation, and financial performance: Analysis ofS&P 500 corporations. Strategic Management Journal, 30:221–231.

Vroom, V. H. 1964. Work and motivation. New York: Wiley.

Weick, K. E. 1979. The social psychology of organizing. Read-ing, MA: Addison-Wesley

Weick, K. E. 1984. Small wins: Redefining the scale of socialproblems. American Psychologist, 39: 40–49.

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

Weick, K. E., Sutcliffe, K.M., & Obstfeld, D. 2005. Organizingand the process of sensemaking. Organization Science,16: 409–421.

Winter, S. G. 2000. The satisficing principle in capabilitylearning. Strategic Management Journal, 21: 981–996.

Winters, D., & Latham, G. P. 1996. The effect of learningversus outcome goals on a simple versus a complextask. Group & Organization Management, 21: 236–250.

Wiseman, R. W., & Catanach, A. H. 1997. A longitudinaldisaggregation of operational risk under changing reg-ulations: Evidence from the savings and loan industry.Academy of Management Journal, 40: 799–830.

Woodman, R. W., Sawyer, J. E., & Griffin, R. W. 1993. Towarda theory of organizational creativity. Academy of Man-agement Review, 18: 293–321.

Young, G., Smith, K., & Grimm, C. 1996. “Austrian” and in-dustrial organization perspectives on firm-level compet-itive activity and performance. Organization Science, 7:243–254.

Sim B. Sitkin ([email protected]) is professor of management, Staudenmeyer Re-search Fellow, and faculty director of the Fuqua/Coach K Center on Leadership andEthics, Fuqua School of Business, Duke University. He received his Ph.D. from StanfordUniversity. His research concerns the influence of leadership and control systems onorganizational and individual change, innovation, trust, learning, risk taking, andaccountability.

Kelly E. See ([email protected]) is assistant professor of management and organi-zations in the Stern School of Business at New York University. She received her Ph.D.in management from Duke University. Her research focuses on the role of learning injudgment under uncertainty and on how factors such as fairness, power, and goalsaffect organizational decision processes.

C. Chet Miller ([email protected]) is the Bauer Professor of OrganizationalStudies at the C. T. Bauer College of Business, University of Houston. He earned hisPh.D. from the University of Texas. His research focuses on upper-echelon leaders,strategic decision processes, and management systems.

Michael W. Lawless ([email protected]) is Tyser Teaching Fellow at the R. H. SmithSchool of Business, University of Maryland. He earned his Ph.D. from the AndersonGraduate School of Management, UCLA. His research takes an evolutionary perspec-tive on technology and innovation management, strategy implementation, and entre-preneurship and new venture financing.

Andrew M. Carton ([email protected]) will be an assistant professor of managementand organization in the Smeal College of Business at The Pennsylvania State Uni-versity beginning in Fall 2011. He received his Ph.D. from Duke University’s FuquaSchool of Business. His research focuses on how leaders adapt to the increasinglycomplex nature of goal systems and intergroup relations in organizations.


Documents

THE PARADOX OF STRETCH GOALS: ORGANIZATIONS IN …performance outcomes as gaps between aspira-tion and current performance elicit and guide effort and persistence (Cyert & March, 1963)