identification of competitor

Copyright # 2002 John Wiley & Sons, Ltd.

MANAGERIAL AND DECISION ECONOMICS

Manage. Decis. Econ. 23: 157–169 (2002)

DOI: 10.1002/mde.1059

Competitor Identification and CompetitorAnalysis: A Broad-BasedManagerial Approach

Mark Bergena,y and Margaret A. Peteraf b,*

aCarlson School of Management, University of Minnesota, 321 19th Avenue South, Minneapolis, MN 55455, USAbTuck School of Business at Dartmouth College, 100 Tuck Hall, Hanover, NH 03755, USA

Managerial myopia in identifying competitive threats is a well-recognized phenomenon(Levitt, 1960; Zajac and Bazerman, 1991). Identifying such threats is particularly

problematic, since they may arise from substitutability on the supply side as well as on the

demand side. Managers who focus only on the product market arena in scanning their

competitive environment may fail to notice threats that are developing due to the resources andlatent capabilities of indirect or potential competitors. This paper brings together insights

from the fields of strategic management and marketing to develop a simple but powerful set of

tools for helping managers overcome this common problem. We present a two-stage

framework for competitor identification and analysis that brings into consideration a broadrange of competitors, including potential competitors, substitutors, and indirect competitors.

Specifically we draw from Peteraf and Bergen’s (2001) framework for competitor

identification to develop a hierarchy of competitor awareness. That is used, in combinationwith resource equivalence, to generate hypotheses on competitive analysis. This framework not

only extends the ken of managers, but also facilitates an assessment of the strategic

opportunities and threats that various competitors represent and allows managers to assess

their significance in relative terms. Copyright # 2002 John Wiley & Sons, Ltd.

INTRODUCTION

Competitor identification is a key task formanagers interested in scanning their competitiveterrain, shoring up their defenses against likelycompetitive incursions, and planning competitiveattack and response strategies. It is a necessaryprecursor to the task of competitor analysis, andthe starting point for analyzing the dynamics ofcompetitive strategy (Smith et al., 1992). Beforeone can assess the relative strengths and weak-nesses of rivals, or track competitive moves andcountermoves, one must first identify the compe-titive set and develop an accurate sense of the

domain in which strategic interactions are likely tooccur.

The purpose of this paper is to provide a set oftractable frameworks for competitor identificationand competitor analysis that facilitate broadenvironmental scanning. To inform our frame-works, we borrow from Peteraf and Bergen’s(2001) framework for competitor analysis. Theirwork borrows from Chen’s (1996) model ofcompetitor analysis, adapting his constructs toour purposes by drawing on the marketingliterature on consumer behavior (Levitt, 1960;Nedungadi, 1990; Peter and Olson, 1993, Mowenand Minor, 1995). Specifically, we bring into sharpfocus the role of customer needs in defining themarketplace to show how a greater recognition ofcustomer needs can expand awareness of whatlurks on the competitive horizon. This allows us toaddress a supply side bias that is often present in

*Correspondence to: Tuck School of Business at Dartmouthcollege, 100 Tuck Hall, Hanover, NH 03755, USA.E-mail: [email protected]@csom.umn.edu

other approaches to competitor identification(Clark and Montgomery, 1999). Moreover, weuse this to develop a hierarchy of competitorawareness that is central to our hypotheses oncompetitor analysis. Further, we introduce theconcept of resource equivalence to facilitate thecomparison of the abilities of indirect andpotential competitors to meet the same set ofcustomer needs as direct competitors. This allowsus to offer a differentiated approach to competitoranalysis.

From a theoretical standpoint, our frameworkscontribute to the development of the stream ofliterature in strategic management on competitivedynamics (Smith et al., 1992; Chen et al., 1992;Chen and MacMillan, 1992; Miller and Chen,1994; Chen and Hambrick, 1995; Ferrier, et al.,1999; Grimm and Smith, 1997) and indeed anchorthis literature at its logical starting point. From apractical standpoint, our approach offers a mark-edly different perspective on competitor identi-fication and analysis, which has importantimplications for managers. Moreover, it providesa mechanism for evaluating competitive threatsand opportunities by comparing firms on the basisof their capabilities to meet market needs.

THE MOTIVATION

Competitor identification serves as an importantfunction in several fields. In industrial organiza-tion economics, it is associated with the task ofdefining markets, which is critical for antitrust andregulatory policy. In marketing, it supports theanalysis of pricing policies, product design, devel-opment and positioning, communications strategy,and channels of distribution. In strategic manage-ment, it provides a foundation for competitoranalysis and the analysis of industry structure,conditions of rivalry, and competitive advantage.

One important objective of competitor identifi-cation is to increase managerial awareness ofcompetitive threats and opportunities. To max-imize awareness, it is essential to survey thecompetitive landscape broadly in the initial stagesof analysis. This can help managers avoid thedangers of a myopic approach to competitivestrategy and will minimize the chance of beingblindsided. It can reduce a firm’s vulnerability tocompetitive blindspots (Zajac and Bazerman,1991), which are particularly troublesome in

settings in which industry boundaries are not welldefined or are very fluid and changeable. Forexample, in emerging industries, in turbulent, highvelocity environments (Eisenhardt and Bourgeois,1989), or in hypercompetitive contexts (D’Aveni,1994), there may be a temptation for managers topay attention only to competitors who display aproduct or technology overlap, because thesecompetitors are salient and because the task ofbroad scanning is difficult. However, it is in thesesettings that competitive encroachments and in-cursions across boundaries by indirect and poten-tial competitors may prove to be the greatestthreat.

A variety of approaches have been developed toaddress the task of competitor identification thatare congruent with market definition. Sophisti-cated quantitative approaches to defining marketsinclude the analyses of cross-price elasticities, ofresidual demand curves, of price correlations, andof trade flows using methods such as the Elzinga–Hogarty approach (Elzinga and Hogarty, 1978).1

Scheffman and Spiller (1987) provide an overviewof classical quantitative approaches to marketdefinition.

Qualitative methods tend to be more ad hoc andare based on the idea that products are in the samemarket if they are close substitutes. Products arejudged to be close substitutes when they are similarin terms of their performance characteristics,occasions for use, and when they are sold in thesame geographic market (Besanko et al., 1996).Qualitative methods derived from economics reston the notion that a market is defined as a ‘set ofsuppliers and demanders whose trading establishesthe price of a good’ (Stigler and Sherwin, 1985).Cognitive methods, in which managers and /orcustomers are queried about which products are incompetition, are more common in the field oforganization theory, which views markets as socialconstructions (See, for example, Porac and Tho-mas, 1990; Porac et al., 1995; Auty and Easton,1990).

Regardless of the analytical approach em-ployed, conceptually it is generally accepted thatcompetitor identification requires the simulta-neous consideration of both demand side andsupply side attributes of putative competitors andtheir domain (Abell, 1980; Day, 1981; Porac andThomas, 1990; Scherer and Ross, 1990; Chen,1996). Demand side considerations ensure thatproducts are substitutable in the eyes of consu-

M. BERGEN ET AL.158

Copyright # 2002 John Wiley & Sons, Ltd. Manage. Decis. Econ. 23: 157–169 (2002)

mers. They include an analysis of the degreeto which products fulfill similar functions andaddress similar needs. Supply side considerationsaddress the degree to which firms are similar interm of technological and production capabilities.

In practice, however, commonly employedapproaches to competitor identification displaybiases toward one side or another. For example, inindustrial organization economics-based ap-proaches, SIC codes (recently replaced byNAICS), have been employed often to delineatemarket boundaries, resulting in a supply side bias(Curran and Goodfellow, 1990). These codes aredetermined on the basis of technological as well asproduct commonalties, which are essentially sup-ply side factors only. Recent experimental evidencesuggests that cognitive views of markets, cham-pioned by Porac and Thomas (1990) as a basis forcompetitor identification, are subject to supplyside biases as well (Clark and Montgomery, 1999).

The problem of bias in market definition is not atrivial one. Awareness is key to organizationalaction and is a principle driver of competitiveattack and response (Chen, 1996). If the approachto market definition is too narrow in scope, thenmanagers may be unaware of activity withcompetitive relevance and may find themselvesblindsided by a surprise attack. Further, they maybe unable to identify attractive market opportu-nities because a narrow view of the market blindsthem to alternative venues in which their capabil-ities may be employed to advantage.

Consider, for example, the irony of the battlethat took place in the 1970s between Polaroid andKodak over the instant camera market (Porter,1983). These two formidable rivals were drawninto a series of enormously costly price wars andlegal battles with one another. While they were soengaged, their markets were being eroded by thearrival of 1-h photo shops, camcorders, andcamera systems that did not require film. Theirfocus on one another as rivals may have preventedthem from seeing the larger picture. And theopportunity cost of their private battle may havebeen the chance to prepare for the greatercompetitive challenge that lay on their horizon.

Few practical tools have been developed toassist managers with the task of identifying andanalyzing competitors that span traditional pro-duct market boundaries. Other approaches tocompetitor identification are congruent with thetask of market definition because they are con-

cerned with identifying only close competitors.This can be limiting, particularly in the contexts ofrapid innovation and of cooperation amongmultimarket firms (Hruska, 1992). Our frameworkgoes well beyond this, since one of its purposes isto maximize managerial awareness of competitiveopportunities and threats. We define competitorsfar more broadly to include not only closecompetitors but more distant competitors in theform of less obvious substitutors and potentialcompetitors as well. In this way, we answer the callof Smith et al. (1992) and Chen (1996) for aframework on competitive dynamics that includespotential competitors.

Moreover, in contrast to most other methods,our approach brings a clear consideration ofcustomer needs into the analysis with respect toboth demand and supply side issues.2 As describedin the next section, we provide a two-stageframework for competitor identification and ana-lysis. In stage 1, we explore how Peteraf andBergen’s (2001) framework can be used to developa hierarchy of competitor awareness that can beused to link competitor identification with compe-titor analysis. This approach addresses the poten-tial supply side bias that often diminishes theeffectiveness of other qualitative approaches (Cur-ran and Goodfellow, 1990; Clark and Montgom-ery, 1999). In the second stage, we provide aframework for competitor analysis that evaluatesand compares the competitors identified in stage 1according to their capabilities for meeting custo-mer needs. We argue that the theoretical content,the practical relevance, and the accessibility of ourframework make it particularly valuable forpractitioners as well as for educators and scholars.

A TWO-STAGE FRAMEWORK

We develop our model in two stages for thefollowing reasons. Firstly, there are two separatetasks to be performed: competitor identificationand competitor analysis. While managers shouldtake a broad approach to competitor identificationto avoid competitive blindspots, they may want tohone in on groups of identified competitorsseparately in performing competitive analysis.Secondly, these two tasks require two separatefunctions. Competitor identification is essentially acategorization task (Rosch, 1978) that involves

COMPETITOR IDENTIFICATION AND ANALYSIS 159


classifying firms on the basis of relevant simila-rities. In contrast, competitor analysis is anevaluative task that goes beyond mere classifica-tion to compare rivals on the basis of relevantdimensions.

Thus, in the first stage, we draw from Peterafand Bergen (2001) to take a broad-based approachto competitor identification, classifying candidatecompetitors on the basis of similarities in terms oftheir resource endowments and the market needsserved. We do so by asking the simple question ofwhether two firms serve the same customer needpresently or have the ability to do so in the nearfuture. The aim of this stage of analysis is to helpmanagers to maximize their awareness of compe-titive threats and to classify the types of competi-tion that they face so that we may develop ahierarchy of competitor awareness that may belinked to competitor analysis.

In the second stage, we take an evaluativeapproach and ask the question of how well twofirms serve the same need or how their capabilitiescompare. We relate that to our framework forcompetitor identification to develop a hierarchy ofcompetitor awareness. We also introduce the notionof resource equivalence to help managers assess thestrengths and weaknesses of their competition interms of comparative capabilities. This takes ourframework into the realm of competitor analysisand allows us to develop a series of propositionsregarding the likelihood of attack and responsefrom different types of competitors.

Stage 1: Recognizing and Classifying the

Competition

To identify and classify the competitive set, wedraw from Peteraf and Bergen (2001) to propose

the framework presented in Figure 1. [see Peterafand Bergen (2001) for the original developmentsand a more detailed discussion of this frameworkfor competitor identification.] We borrow fromChen’s (1996) highly acclaimed paper, but adapthis constructs of market commonality and re-source similarity to serve our different purposes.Specifically, under the category of market com-monality, we sort competitors based on the degreeto which they address similar customer needs,while under the category of resource similarity, wesort competitors based on the degree to whichtheir resource endowment is similar in terms oftype or composition.

Chen (1996, p.106) defines market commonalityas ‘the degree of presence that a competitormanifests in the markets it overlaps with the focalfirm’, a definition that serves as an indicator of afirm’s ‘direct or primary competitors’ and theirbehavior (Chen, 1996 p.102). We broaden thedefinition to take the perspective that firmscompete with one another to the extent that theysatisfy the same customer needs. Accordingly, weredefine market commonalty as the degree to which

a given competitor overlaps with the focal firm in

terms of customer needs served. This approach isconsistent with the marketing literature (Levitt,1960; Cooper and Inoue, 1996) and recognizes thatcompetitors may include firms that do not sharethe same technological platform. It is consistent aswell with the type of approach to market or nicheoverlap utilized in the population ecology litera-ture (McPherson, 1983; Baum and Singh, 1994;Baum and Korn, 1996).3

Notice how the incorporation of customer needschanges awareness through a change in theassessment of the competitive structure. Forinstance, in analyzing the airline industry, acustomer-needs-based analysis would suggest thatthe fundamental enduring need is for convenienttransportation. In short, haul airline markets,there are a number of viable alternatives forconsumers, including rail, automobile, bus, andlimousine services. Similarly, Southwest Airlines’entry into the Northeast corridor through Islipposed a significant competitive threat to Amtrak.Our needs-based definition encourages managersto look beyond restrictive product market bound-aries to assess competitive threats more broadly. Itcaptures the extent to which direct and indirectcompetitors, such as substitutors, are in competi-tion to serve the same needs.

Resource Similarity

Mar

ket

Com

mon

alit

y Direct CompetitorsIndirect Competitors

(Substitutes)

Potential Competitors

Figure 1. Mapping the competitive terrain.

M. BERGEN ET AL.160


Chen (1996, p.107) defines resource similarity as‘the extent to which a given competitor possessesstrategic endowments comparable, in terms ofboth type and amount, to those of the focal firm’.We modify this definition slightly. Because wemerely categorize firms by type of competitor instage 1, reserving comparisons regarding relativecompetitive capabilities for stage 2 analysis, we donot consider resource amounts. We regard theamount of resources as one of the many dimen-sions on which a firm may claim resource super-iority relative to other firms. For our purposes, weredefine resource similarity as the extent to which a

given competitor possesses strategic endowments

comparable, in terms of type, to those of the focal

firm.

We utilize these two constructs of marketcommonality and resource similarity to categorizethe competitive field from the point of view of afocal firm. Thus we employ dyadic comparisonsbetween that firm and candidate rivals. Bydisplaying resource equivalence as an increasingfunction on the x-axis and market commonality asan increasing function on the y-axis, we can mapthe competitive field of a focal firm by locatingcandidate rivals on the grid.

A firm that scores high in terms of both marketcommonality and resource similarity is one thatserves the same market needs with the same typesof resources as the focal firm. Firms such as thesewill be found in the northeast corner of the grid.These are the focal firm’s direct competitors, as forexample AMD is for Intel (both producers ofmicroprocessors and suppliers to computer man-ufacturers).

Firms with resource endowments similar to thefocal firm that do not presently serve the samecustomer needs will be found in the southeastcorner of the grid. These are the set of potential

entrants into the markets of the focal firm. Forexample, a caterer and a local restaurant may bothcompete on the basis of their reputations for goodfood and service, and similarly employ chefs,kitchen equipment, and the like. Although theirresource similarity is high, they may neverthelesscater to demonstrably different customer needs.Corporate customers need caterers to prepare,deliver, and serve party foods and dinners for largefunctions held at the client’s site; individualspatronize restaurants in small groups for apleasurable dining experience away from homeor for small-scale take-out service. While these two

businesses presently serve different needs, eithercould choose to branch out and enter the others’market since they already have most of theresources required to do so.

Firms that occupy the southwest corner of thegrid score low on both dimensions. They areentirely outside the competitive set at present,although this could change over time as firmschange their positions. Of greater interest is the setof firms in the northwest corner. These firms areserving the same market needs as the focal firm butwith different types of resources. They comprisethe set of indirect competitors, such as substitutors.

Substitutors are an important, but often invi-sible, class of competitors since they frequentlyutilize new technologies, whose costs are likely todecline due to the learning curve. For example,cameras may be used to take pictures with film-based technologies that depend on capabilities inchemistry and mechanics. Alternatively, digitalpictures may be produced using electronics cap-abilities. The underlying consumer need (to recordevents pictorially) is served equally well bycompletely different technologies, as a result ofwhich Sony and Kodak now compete (Friedman,1999).

While this mapping exercise is fairly straightfor-ward, our customer needs perspective introducesan important subtlety into the analysis thatmanagers might otherwise overlook. Consideragain the set of direct competitors. When con-sumers acquire information about available alter-natives to fulfill a need, they generate an‘awareness set’ that is subsequently honed downto make the choice problem easier (Peter andOlson, 1993). This reduced set, known in the fieldof marketing as the ‘consideration set’, is the set ofoptions from which the eventual choice generallyemerges. To the extent that more than onealternative is actively considered in the con-sideration set, those alternatives are the mostsignificant competitors. In other words, the directcompetitors identified by the mapping exercisemay not all be immediate competitive threats,because only a subset of them may be inconsumers’ consideration sets. This suggests thatthe needs perspective can be used both broadlyand more narrowly to generate a deep under-standing of the variations and ambiguities in thecompetitive landscape.

There is an important strategic implication ofthis line of thinking. Consideration sets do not



exist in a vacuum. In fact, firms frequently attemptto influence the composition of the considerationset through a multitude of devices includingcomparative advertising, sales presentations, andshelf placements. Industry leaders attempt toassure that they are the sole members of theconsideration set, and therefore rarely engage incomparative advertising that identifies a compe-titor. Conversely, firms that have weaker marketpositions attempt to develop psychological asso-ciations with industry leaders through compara-tive ads that specifically identify the principalcompetitor. Such associations allow weak firms toinveigle their way into customers’ considerationsets.

This framework, then, is useful not only forincreasing awareness of the various dimensions ofthe competitive landscape, but it can assist man-agers with attempts to influence the composition ofthe landscape as well. It is useful for identifyingopportunities for collaborative and cooperativeactivities, such as joint advertising to increaseindustry demand, in addition to identifying andmonitoring threats. Additionally, there are otherpractical implications of this framework. It can beused to view competition dynamically and totrack potential competitors’ movements over time.It serves to remind managers to track not onlyrivals’ conduct in product markets, but theiractivities in factor markets as well. Activity in theresource market may provide a forewarning ofimpending competitive action in the productmarket.

The limitation to this part of the framework isthat while it is useful for surveying the competitiveterrain and classifying competitors, it cannot beused to order the terrain and rank competitivethreats and opportunities. That is to say, it cannotbe used to ascertain which competitors present thestrongest threat to a focal firm and which are themost vulnerable to competitive attack. For this, weneed additional information. Specifically, we needto be able to evaluate the differences in resourcetype, in terms of their abilities to satisfy a given setof market needs.4 Knowing that the capa-bilities of two players differ is useful for simpleclassification purposes. But in order to predictwhich is the stronger competitor, we need toknow how their capabilities differ and which set isbetter suited to the market needs being served. Weturn to these issues in the second stage of ouranalysis.

Stage 2: Evaluating the Competition and

Predicting Rivalry

To facilitate evaluating the competition andmaking predictions regarding the likelihood ofattack and response, we introduce the construct ofresource equivalence. We define resource equiva-lence as the extent to which a given competitor

possesses strategic endowments capable of satisfying

the same customer needs as the focal firm. Althoughthis is a continuous measure like the notion ofresource similarity, we can conceive of it in termsof high and low degrees. If two firms have highresource equivalence, they come close to beingequally capable of satisfying the same customerneeds. Equal capability implies that they do or canaddress the same market needs equally well. Forexample, among supermarket chains, Kroger andAlbertsons score high in terms of resourceequivalence, and represent relatively balancedcompetitors nationwide, with approximately 1900stores apiece in 1999.5

We utilize this construct to conduct competitoranalysis by assessing the strength of various typesof competitors relative to a focal firm, as weillustrate in Figure 2.

On the horizontal axis, we sort firms accordingto whether they have high-or low-resource equiva-lence relative to the focal firm. On the vertical axis,we sort firms according to the three basiccompetitive categories identified in the first stageof the analysis: direct competitors, potentialcompetitors, and indirect competitors.

Note that there is relative balance in terms ofcompetitive strength between the focal firm and itscompetitors on the right-hand side of matrix. Forexample, if the lower right-hand side of the matrix

IndirectCompetitors

e.g. Peapod vs. Albertsons e.g. Wal-Mart vs.Albertsons

PotentialCompetitors

e.g. Canadian Safeway vs.Albertsons

(e.g. none)

DirectCompetitors

e.g. Lunds/Byerly's vs.Albertsons

e.g. Kroger vs. Albertsons

Low High

Resource Equivalence

Figure 2. A framework for competitor analysis.

M. BERGEN ET AL.162


contains Kroger (in relation to Albertsons), theupper right cell might contain Wal-Mart’s Super-centers, which presently are only a third of the sizeof the traditional supermarket chains (with lowerscale and scope economies), but have superiorcapabilities in terms of information technologyand logistics (see Note 5). It is a balanced packagesuch that their set of capabilities, takenas a whole at this point in time, can meet customerneeds about as well as the capabilities of Albert-sons. The middle cell on the right might containthe large Canadian chain, Loblaw, that presentlyserves other geographic needs but with similar andequivalent capabilities.

On the left-hand side of the matrix, there isimbalance with respect to capabilities, implyingthat one rival is stronger than the other. Forexample, a small chain such as Lunds/Byerly’swith its 19 stores would occupy the lower left cell,while middle-sized Canada Safeway might occupythe middle cell, and a small but growing Internetprovider, such as Peapod, may be found in theupper left cell (see Note 5). At present, none ofthese competitors is a match for the superiorcapabilities of Albertsons to meet customer needs,although that could change over time. Note, moregenerally, that whereas this matrix is indicative ofcompetitive imbalance, it does not reveal thedirectionality of the imbalance. Individual compe-titors occupying the left-hand side could bestronger or weaker than the focal firm.

We capture the relationship between competi-tive balance and resource equivalence with thefollowing premise:

Premise 1:

As resource equivalence increases between a focalfirm and a rival, the degree of competitive balanceincreases as well, Ceteris paribus.

While balance is the indicator that increasesfrom left to right on this diagram, it is awarenessthat increases as we move from the bottom to thetop. The reason for this is simple. Direct compe-titors have the most in common with the focalfirm, since there is similarity in terms of bothcustomer needs met and resource type. (This ofcourse holds the level of resource equivalenceconstant.) They are uppermost in the minds ofmanagers because of the salience of these types ofsimilarities. The work by Reger and Huff(1993), and Lant and Baum (1995), Porac andBaden-Fuller (1989) on how managers perceive

competitors provides empirical support for theseclaims.

Potential competitors have similar resources asthe focal firm but serve different market needs.Indirect competitors represent the reverse situa-tion. Because of the supply side bias that wediscussed in the previous section, managers paygreater attention to firms with similar technologiesand resources when scanning their competitiveenvironment than they do to firms that are outsidetraditional product market boundaries. For thesereasons, we suggest the following:

Premise 2:

The awareness between a focal firm and its rivalsdecreases, Ceteris paribus, as we move from theclass of direct competitors, to potential competi-tors, to indirect competitors.

Our analysis of the degree of rivalry in varioussegments of the competitive domain depends on anassessment of the two factors of balance andawareness. An understanding of the level ofawareness between competitors is essential sinceawareness is one of the three fundamental driversof competitive behavior, which also includemotivation and capability.6 As Chen (1996) makesclear, organizational action depends criticallyupon these three drivers. An understanding ofthe degree of competitive balance is importantbecause it affects both capability, which is arelative phenomenon, and motivation, whichdepends in part upon the probability of success.For example, when there is balance, then thecapability of one party to dominate in a compe-titive battle is low relative to the situation whenthere is imbalance. In this situation, there is atendency toward what is known in the literature onmulti-market competition as ‘mutual forbearance’(Gimeno and Woo, 1996,1999; Baum and Korn,1996, 1999; McGrath et al., 1998). Balance alsoaffects the motivation of competitors in thatmotivation is affected, in part, by the likelihoodof success. If the level of balance is high, forexample, then the probability of successfullydefeating one’s rival is low relative to an imbal-anced situation in which one holds the advantage.

We use these premises to derive a series oflogical propositions regarding the likelihood ofattack and response from various competitivequarters. The most general of these follow fromthe effect of balance on rivalry. Balance in-hibits the initiation of competitive action, since



competitors have more to gain from forbearancethan from competitive engagement. That is, theinitiator’s probability of success is lower and itsprobability of defeat is higher in a balancedsituation. In addition, balance often encouragescooperative actions that may provide a win–winsolution for the parties involved (Brandenburgerand Nalebuff, 1996). When there is imbalance interms of capabilities, advantaged rivals are moti-vated to capitalize on their situation, striking fromstrength against weakness, or using a judo strategyto turn an opponent’s strength into a liability(Gelman and Salop, 1983; Yoffee and Cusumano,1999). In summary, when the level of balance ishigh, the level of motivation for competitive attackis low. Because of the corresponding relationshipbetween balance and resource equivalence, wepropose the following:

Proposition 1:

As the resource equivalence between two firmsincreases, the likelihood of attack is reduced,Ceteris paribus.

Because there is greater balance on the right-hand side of the figure, the likelihood of responseto attack is also greater on the right-hand side. Thereason for this is that the capability of effectiveresponse is much higher when resource equiva-lence is high and the motivation is higher as welldue to the higher probability of success. This, ofcourse, assumes that other things that affectmotivation, such as the gains from engaging incompetitive action, are equal across conditions.Notice, parenthetically, that this high likelihood ofa response to an attack is one of the things thatmotivate forbearance from attack in the first place.This reasoning leads us to our next proposition:

Proposition 2:

As the resource equivalence between two firmsincreases, the likelihood of response to a compe-titive attack increases, Ceteris paribus.

This proposition reinforces our third proposi-tion, that rivalry is reduced among resourceequivalent competitors. The reason for this argu-ment is that firms will be even less likely to upsetthe competitive balance and attack if they knowthat their equally capable rival is likely to respond.They are better off accepting an uneasy truce thanrisking a costly and damaging competitive battle.

Along the two sides of the matrix, the propertiesof awareness and balance interact in interestingways to yield further predictions regarding thelikelihood of attack and response. Note first thatalthough the likelihood of attack is generallylowest on the right-hand side of the diagram, itincreases as we move from the bottom to the topcell. The reason for this is that as awareness ofrivals decreases, the awareness of the balancedecreases as well. Notice that it is more difficult toperceive resource equivalence when the capabilitiesdiffer by type and by composition, as in the Wal-Mart versus Albertsons example that we havealready given. Thus, balance is less of alimiting factor in the uppermost cells on rivalryand opportunities for mutual understandingdiminish due to the greater heterogeneity. More-over, an aware rival may take advantageof the fact that on average, his competitive targetis likely to be relatively unaware of competitiveactions coming from an indirect competitor. Thisincreases the probability of a successful assaultand thus the motivation for action. In sum, wepropose:

Proposition 3:

When the degree of resource equivalence is high,the likelihood of attack increases as we move fromthe class of direct competitors, to potentialcompetitors, to indirect competitors.

The likelihood of response runs in the oppositeorder on the left-hand side. That is, balanceddirect competitors are the most likely to respondto an attack, while indirect competitors are theleast likely to respond. When there is balance interms of resource equivalence, the capability foreffective response is high. Because the motivationfor action is dependent on an assessment of thelikelihood of success, the motivation for responseis high as well, other things equal (such as what isat stake). Even though there may be equivalentcapabilities among indirect competitors, they areless likely to be perceived as equivalent. Moreover,rivals are likely to be less aware that indirectcompetitors are indeed competitors, and thisreduced awareness should be accompanied by areduction in responsiveness.

Proposition 4:

When the degree of resource equivalence is high,the likelihood of response decreases as we move

M. BERGEN ET AL.164


from the class of direct competitors, to potentialcompetitors, to indirect competitors.

The action and response expectations on theleft-hand side of the matrix are just the reverse ofthose on the right. Thus, the likelihood of attack isgreatest when imbalanced rivals compete directlyand lowest when imbalanced rivals competeindirectly. The reason for this is that the awarenessof a rivalrous opportunity will be greater amongdirect competitors than among potential competi-tors or indirect competitors. The propensity ofmanagers to view rivals in terms of similarproducts along traditional market boundary linesmeans that they will choose opportunities tocompete first in those venues. In addition, theymay feel that they know better how to competeeffectively within traditional market boundariesthan across them. By remaining within theseboundaries, they can utilize available organiza-tional routines for competing locally and have arepository of knowledge regarding likely reactionsand outcomes that are most applicable for suchactions (Nelson and Winter, 1982). In sum:

Proposition 5:

When the degree of resource equivalence is low,the likelihood of attack decreases as we move fromthe class of direct competitors, to potentialcompetitors, to indirect competitors.

Here the outcome is opposite for the likelihoodof response. To understand this, one must firstappreciate that when there is imbalance amongrivals, the likelihood of effective response is small.This reduces the likelihood of response since bothcapability and motivation will be low, other thingsequal. What drives a lower response rate amongindirect competitors relative to direct competitorson this side of the matrix is the lower awareness ofthe imbalance and its implications. That is, whileindirect competitors are unlikely to see each otheras rivals a priori, when attacked they are morelikely to respond simply because the perception ofimbalance is lower. They are more likely to believethat they have a chance of effective response,Ceteris paribus. This leads to the followingprediction:

Proposition 6:

When the degree of resource equivalence is low,the likelihood of response increases as we move

from the class of direct competitors, to potentialcompetitors, to indirect competitors.

We now turn to a discussion of the uses of ourframework and its limitations in the sectionbelow.

DISCUSSION

Implications

Our frameworks can be helpful to strategists in avariety of ways. First, the stage 1 framework canbe used in isolation as an aid to overcoming anatural tendency toward over weighting supplyside factors in surveying the competitive environ-ment and ignoring competition from beyondtraditional product market boundaries. Second, itcan be used to chart the movement of competitorsto new positions along the grid. Thus, it canprovide a more dynamic outlook of how thecompetitive situation is changing. Third, it can beused to look for new opportunities for competitivedominance. It can also be used to search forcooperative opportunities. Fourth, it can be usedto design a strategy to influence customer needs, aswell as their awareness sets and considerationssets. Thus, it can be used to change the competitivelandscape on a variety of dimensions.

Stage 2 of the analysis has significant implica-tions for managers as well. Not only do we believethat our propositions have predictive powerregarding the likelihood of attack and response,but in addition, they can be used to order thecompetitive field in terms of threats and opportu-nities. For example, on the right-hand side of thematrix, it is clear that the degree of competitivethreat increases from bottom to top if we view thecompetitive field dynamically. That is, over timethe indirect competitors among those competitorswith equivalent resources pose the greatest threatsto a focal firm. This is consistent with Chen’sprediction that the greatest competitive threatcomes from rivals with low market commonalityand high resource similarity. Applying this to ourstage 1 framework suggests that potential compe-titors (with low market commonality and highresource similarity) may ultimately be morethreatening than direct competitors (with highmarket commonality and high resource similarity),which accords with our predictions.



Because stage 2 matrix indicates competitiveimbalance, but does not show whether it is therival or the focal firm that benefits from theimbalance, the left-hand side of the matrix revealsinformation about both opportunities and threats.If the focal firm is the holder of resources that areless capable of meeting market needs than its rival,then its greatest threat (over time) will be foundamong its direct competitors, since the propensityfor attack is strongest and the propensity forresponse is weakest. If the focal firm is theadvantaged party, then this cell presents its great-est opportunities for successful competitive en-gagement. More generally, the matrix revealsopportunities for cooperation, along the right-hand side and suggests that a firm seek competitiveopportunities and threats along the left-hand side.Note that if the focal firm is less capable ofmeeting market needs than stronger rivals, itsgreatest competitive threat will come from thedirect competitors in the lower left cell. These willcomprise the competitive set that pose the mostdanger since the focal firm will be least able torespond effectively.7 Further, it is in this cell thatthere is the least opportunity for cooperation.Cooperative opportunities are greatest on theright-hand side of the matrix. In addition, it ispossible that relative capability among rivals iseven more important in predicting rivalrousinteractions than the awareness factor. While itmay not be a more significant determinant in itsown right, its effect is magnified since it affectsmotivation as well through its effect on thelikelihood of success.

There are other managerial implications as well.Since the competitive threat is relatively high fromindirect competitors with resource equivalence,managers should actively seek opportunities tofind stable win–win solutions. An example of thisis the recent set of agreements reached betweenMicrosoft and Apple. Since the primary driver ofincreased likelihood of attack on this side of thematrix is reduced awareness of the competitivebalance, managers should actively seek to educatetheir rivals regarding their resource equivalence.They should also seek to signal both theircapability of an effective response to any attackand their willingness to respond in order todiminish further the probability of attack fromall types of resource equivalent competitors.

When resources are not equivalent and the focalfirm is in the superior position, managers should

take steps to enhance awareness of their presence,so that rivals will recognize the futility of theirposition and refrain from response. If the focalfirm is in the weaker position, its managers shouldseek to diminish awareness of this imbalance, so asnot to invite an attack, or try to commit crediblyto staying weak. They might also consider signal-ing that they believe that their capabilities arestronger than they are or that they will respondirrationally to an attack, in order to diminishattack probabilities. Finally, just as managers canuse the stage 1 competitive terrain map to trackcompetitive movements, so can they use stage 2matrix to chart changes in resource equivalenceand corresponding capabilities. They can use thisinformation to prepare defenses against competi-tive incursions and to strengthen their position vis-a-vis rivals. For example, grocery chains such asAlbertsons and Kroger have begun to consolidateto improve their capability to meet market needsefficiently as they have noticed Wal-Mart gainingstrength with their hypermarkets and supercenters.

Limitations

One limitation of our model is that, in contrast toChen’s (1996) model, it does not account for therelative importance of various market segments tothe industry participants. Thus some marketsegments may be of greater strategic value to firmsthan others and may be worth fighting for morethan others. This affects rivalry through its effecton motivation. We have assumed this effect to beconstant throughout our analysis. Managersshould be aware of this limitation and shouldfactor in the relative importance of various marketsegments to their own firms and to rivals inmaking judgments regarding likely competitivereactions. In addition, managers should considerthe growth rate of various market segments, sincethat will define the attractiveness of the turf undercontention and will affect the level of rivalry(Porter, 1980).

Another limitation is that the model does nottake into account the notion that some resourcesrepresent more potent competitive weapons thanothers and may also make rivals more immune tocompetitive attack. Our model addresses thequestions of which firms are alike and which aresuperior, but not which are assailable. To anextent, these factors are built into our assumptionsregarding the meaning of resource equivalence, but

M. BERGEN ET AL.166


the underlying micro-level analysis is beyond thescope of this paper. By incorporating additionalinformation regarding what drives competitiveadvantage from models such as the resource basedview of the firm (Barney, 1991; Amit and Schoe-maker, 1993; Peteraf, 1993), it is conceivable thatthe power of this model could be increased. Thismay be a fruitful avenue for future research.

The model is limited as well, in that it ignoresthe competition that can come from suppliers andcustomers, as Porter’s (1980) five-forces modelsuggests. Neither does it take into account otherforms of competition within a firm’s ‘value net’(Brandenburger and Nalebuff, 1996). Moreover,while the model has the capability to consider therole of complementors, we do not explicitly discussthis issue. Since complementors play a cooperativerole more often than a competitive role, includingthem in stage 2 analysis may be misleading.

In addition, while the model can be used to anextent to track competitive movement, it lacks therichness of a truly dynamic model. It ignoresfactors of timing, such as the length of responsedelay, subsuming these issues instead into the‘likelihood’ of response. Additional insight fromthe model might be possible if one built someconsideration of the relative adjustment costs thatvarious rivals face in altering their capability sets.See, for example, the Dierickx and Cools’ (1989)discussion of capabilities and time compressiondiseconomies. See also the Grimm and Smith’s(1997) discussion of response delays and othertiming issues.

Finally, it is possible that the model’s predictivepower could be increased if it accommodated inexplicit fashion an understanding of how thecapabilities of rivals create value in terms ofincreasing perceived customer benefits and low-ering costs. By delving more deeply into theseaspects, one could predict even more cleanly wherea firm’s best opportunities and greatest threats lie.We reserve these possibilities for future work.

CONCLUDING REMARKS

This paper makes its contribution to scholarshipon four broad fronts. First, we extend thecompetitive dynamics literature to include the taskof competitor identification. We do so in a waythat is consistent with and complementary to the

thinking in this research stream, facilitating seam-less integration across the analytic tasks andcontributing to a more complete overall model ofcompetitive dynamics. Second, we focus attentionon the role of the customer in defining competitorsand show how a greater consideration of customerneeds can expand managerial awareness of whatlurks on the competitive horizon. Third, weintroduce the notion of resource equivalence as atool for evaluating competitors. This is a powerfulconstruct that directs attention to competitivedimensions that matter at a fundamental level.Fourth, we use our hierarchy of competitorawareness and resource equivalence to generatehypotheses on competitive analysis.

The paper’s contributions are not limited,however, to the world of academia. The frame-works and propositions in this paper also havestrong implications for managers and for strate-gists. The paper provides a set of tools forscanning the competitive horizon that can helpmanagers to lessen the probability of their beingblindsided by a competitive attack. These tools canhelp managers take a more dynamic view ofcompetition and think more strategically aboutcompetitive challenges that lie ahead. With agreater appreciation of the competitive dynamicsof the broad environment in which they operate,they may become more proactive in responding tothese challenges. They can also help managers toanticipate new opportunities that others might notsee so readily, thus enabling them to move morequickly and take advantage of them as they arise.Finally, the tools for competitor analysis providedin this paper can also help managers to gain adeeper appreciation of what drives attack andresponse behavior. This will enable them toanticipate their rival’s moves more accuratelyand to respond more rationally. As managerslearn to navigate their competitive terrain withgreater competency, they should see improvementsin both the competitive position and performanceof their firms, relative to more shortsightedrivals.

Acknowledgements

The authors thank Ming-Jer Chen, Wally Ferrier, PaulWolfson, Luohua Zhou, and Zeev Rotem for helpful discus-sions. We are grateful to Sourav Ray for his research assistance.In addition, we thank Orv Walker and his marketing strategyseminar participants at the University of Minnesota, as well asone anonymous reviewer. Special thanks to Akshay Rao for hisfundamental contributions to this paper.



NOTES

1. See Besanko, Dranove, and Shanley (1996) for adescription of these techniques.

2. Chen (1996) also provides a model of competitoranalysis that accommodates a broad approach to themarket. Our framework is differentiated from his inthat it addresses explicitly the task of competitoridentification (while his takes the market boundariesas given). In addition, it incorporates attending tocustomer needs as a basis for both identifyingmarkets and comparing capabilities.

3. The organizational niche overlap concept is in turnrelated to the concept of domain similarity in theliterature on interorganizational relationships. Itrefers to the extent to which firms obtain revenuesfrom the same sources, have similar skill bases, andprovide similar products and services (Baum andSingh, 1994).

4. More specifically, we are concerned with relativeabilities to create value in a specific context.

5. From data compiled by the Retail Food IndustryCenter of the University of Minnesota.

6. These drivers are drawn from the literature onchange, learning, and decision-making. See, forexample, Schelling (1960), Allison (1971), Kieslerand Sproull (1982), Dutton and Jackson (1987),Lant, Milliken, and Batra (1992).

7. This is true in a static sense as well as a dynamicsense, since focal firms of this sort are not likely tolast long unless their capabilities improve.

REFERENCES

Abell DF. 1980. Defining the Business: Starting Point ofStrategic Planning. Englewood Cliffs, NJ: Prentice-Hall.

Allison GT. 1971. Essence of Decision: Explaining theCuban Missile Crisis. Boston: Little, Brown.

Amit R, Schoemaker PJ. 1993. Strategic assets andorganizational rent. Strategic Management Journal 14:33–46.

Auty S, Easton G. 1990. Patterns of competition: astudy of local restaurants. Marketing Intelligence &Planning 8: 27–34.

Barney JB. 1991: Firm resources and sustained compe-titive advantage. Journal of Management 17: 99–120.

Baum JA, Korn HJ. 1996. Competitive dynamics ofinterfirm rivalry. Academy of Management Journal 39:255–291.

Baum JA, Korn HJ. 1999. Dynamics of dyadiccompetitive interaction. Strategic Management Jour-nal 20: 251–278.

Baum JA, Singh JV. 1994. Organizational niche overlapand the dynamics of organizational founding. Orga-nization Science 5: 483–501.

Besanko D, Dranove D, Shanley M. 1996. TheEconomics of Strategy. New York: Wiley.

Brandenburger AM, Nalebuff B.J. 1996. Co-opetition.New York, NY: Doubleday.

Brandenburger AM, Stuart HW. 1996. Value-BasedBusiness Strategy. Journal of Economics and BusinessStrategy 5: 5–24.

Chen M.-J. 1996. Competitor analysis and interfirmrivalry: toward a theoretical integration. Academy ofManagement Review 21: 100–134.

Chen M.-J, Hambrick DC. 1995. Speed, stealth, andselective attack: how small firms differ from largefirms in competitive behavior. Academy of Manage-ment Journal 38: 453–482.

Chen M.-J, MacMillan IC. 1992. Nonresponse anddelayed response to competitive moves: The roles ofcompetitor dependence and action irreversibility.Academy of Management Journal 35: 539–570.

Chen M.-J, Smith KG, Grimm CM. 1992. Actioncharacteristics as predictors of competitive responses.Management Science 38: 439–455.

Clark BH, Montgomery DB. 1999. Managerial Identi-fication of Competitors. Working Paper No. 98–127,Marketing Science Institute, Cambridge, MA.

Cooper LG, Inoue A. 1996. Building market structuresfrom consumer preferences. Journal of MarketingResearch 33: 293–306.

Curran JG, Goodfellow JH. 1990. Theoretical andpractical issues in the determination of marketboundaries. European Journal of Marketing 24:16–29.

D’Aveni R. 1994. Hypercompetition. New York: FreePress.

Day GS. 1981. Strategic market analysis and definition:an integrated approach. Strategic Management Jour-nal 2: 281–299.

Dierickx I, Cool K. 1989. Asset stock accumulation andsustainability of competitive advantage. ManagementScience 35: 1504–1511.

Dutton JE, Jackson SB. 1987. Categorizing strategicissues: links to organizational action. Academy ofManagement Review 12: 76–90.

Eisenhardt K, Bourgeois J. 1989. Politics of strategicdecision making in high-velocity environments: To-ward a midrange theory. Academy of ManagementJournal 31: 737–770.

Elzinga K, Hogarty T. 1978. The problem of geographicmarket definition revisited: the case of coal. AntitrustBulletin 23: 1–18.

Ferrier WJ, Smith KG, Grimm CM. 1999. The role ofcompetitive action in market share erosion andindustry dethronement: a study of industry leadersand challengers. Academy of Management Journal 42:372–389.

Friedman TL. 1999. The Lexus and the Olive Tree. NewYork: Farrar, Straus & Giroux.

Gelman J, Salop S. 1983. Judo Economics: capacitylimitation and coupon competition. Rand Journal ofEconomics (Autumn): 315–325.

Gimeno J, Woo CY. 1996. Hypercompetition in amultimarket environment. The role of strategicsimilarity and multimarket contact on competitivede-escalation. Organization Science 7: 322–341.

Gimeno J, Woo CY. 1999. Multimarket contact,economies of scope, and firm performance. Academyof Management Journal 42: 239–259.

M. BERGEN ET AL.168


Grimm CM, Smith KG. 1997. Strategy as Action:Industry Rivalry and Coordination. Cincinnati: South-Western College Publishing.

Hruska AC. 1992. A broad market approach to antitrustproduct market definition in innovative industries.The Yale Law Journal 102: 305–342.

Kiesler S, Sproull L. 1982. Managerial response tochanging environments: perspectives on problemsensing from social cognition. Administrative ScienceQuarterly 27: 548–570.

Lant TK, Baum JA. 1995. Cognitive sources of sociallyconstructed competitive groups: examples from theManhattan hotel industry. In The Institutional Con-struction of Organizations, Scott WR, Christensen S.(eds). Thousand Oaks, CA: Sage.

Lant TK, Millikin FJ, Batra B. 1992. The role ofmanagerial learning and interpretation in strategicpersistence and reorientation. Strategic ManagementJournal 13: 585–608.

Levitt T. 1960. Marketing myopia. Harvard BusinessReview 38: 45–56.

McGrath RG, Chen M.-J, MacMillan IC. 1998. Multi-market maneuvering in uncertain spheres of influence:Resource diversion strategies. Academy of Manage-ment Review 23: 724–740.

McPherson JM. 1983. An ecology of affiliation. Amer-ican Sociological Review 48: 519–532.

Miller D, Chen M.-J. 1994. Sources and consequences ofcompetitive inertia: a study of the US airline industry.Administrative Science Quarterly 39: 1–23.

Mowen JC, Minor M. 1995. Consumer Behavior, NewJersey: Prentice-Hall.

Nedungadi P. (1990). ‘Recall and consumer considera-tion sets: influencing choice without altering brandevaluations.’ Journal of Consumer Research 17(3):263–276.

Nelson R, Winter S. 1982. An Evolutionary Theoryof Economic Change. Cambridge, MA: BelknapPress.

Peter JP, Olson JC. (1993). Consumer Behavior andMarketing Strategy. Boston: Irwin.

Peteraf MA. 1993. The cornerstones of competitiveadvantage: a resource-based view. Strategic Manage-ment Journal 14: 179–191.

Peteraf MA, Bergen M. 2001. Scanning DynamicCompetitive Landscapes: A customer and resourcebased framework. Working paper, Carlson School ofManagement, University of Minnesota.

Porac JF, Baden-Fuller C. 1989. Competitive groups ascognitive communities: the case of Scottish knitwearmanufacturers. Journal of Management Studies 26:397–416.

Porac JF, Thomas H. 1990. Taxonomic mental modelsin competitor definition. Academy of ManagementReview 15: 224–240.

Porac JF, Thomas H, Wilson F, Paton D, Kanfer A.1995. Rivalry and the industry model of Scottishknitwear producers. Administrative Science Quarterly40: 203–228.

Porter ME. 1980. Competitive Strategy: Techniques forAnalyzing Industries and Competitors. New York:Free Press.

Porter ME. 1983. Cases in Competitive Strategy. NewYork: Free Press.

Reger RK, Huff AS. 1993. Strategic groups: a cognitiveperspective. Strategic Management Journal 7: 485–501.

Rosch E. 1978. Principles of categorization. In Cognitionand Categorization. Rosch E, Lloyd B. (eds). Hillsdale,NJ: Erlbaum.

Scheffman DT, Spiller PT. 1987. Geographic marketdefinition under the US Department of Justice mergerguidelines. Journal of Law and Economics 30: 123–148.

Schelling TC. 1960. The Strategy of Conflict. Cam-bridge, MA: Harvard University Press.

Scherer FM, Ross D. 1990. Industrial Market Structureand Economic Performance. Boston: Houghton Mif-flin.

Smith KG, Grimm CM, Gannon MJ. 1992. Dynamics ofCompetitive Strategy. Newbury Park, CA: Sage.

Stigler G, Sherwin R. 1985. The extent of the market.Journal of Law and Economics 28: 555–585.

Yoffee D, Cusumano M. 1999. Judo strategy: Thecompetitive dynamics of Internet time. HarvardBusiness Review January–February: 70–81.

Zajac EJ, Bazerman MH. 1991. Blind spots in industryand competitor analysis: implications of interfirm(mis)perception to strategic decisions. Academy ofManagement Review 16: 37–46.



identification of competitor

Documents