Look Before You Leap: Market Opportunity Identification in Emerging Technology Firms

Abstract

Entrepreneurs play a fundamental role in bringing new technologies to market. Because technologies are often configurable to serve a variety of different markets, it is possible for entrepreneurs to identify multiple market opportunities prior to the first market entry of their emerging firms, and if they elect to do so, to therefore have a choice of which market to enter first. The empirical results presented in this paper offer three new insights regarding this important early-stage choice in new firm creation. First, they reveal that serial entrepreneurs have learned through prior start-up experience to generate a "choice set" of alternative market opportunities before deciding which one to pursue in their new firm creation. Second, the analysis indicates that entrepreneurs who identify a "choice set" of market opportunities prior to first entry derive performance benefits by doing so. Third, the positive relationship between the number of market opportunities identified prior to first entry and new firm performance is nonlinear and subject to decreasing marginal return. The research literature has yet to acknowledge the notion of multiple opportunity identification prior to entry, and the related idea of selecting the most favorable market opportunity for the creation of a new technology firm.

Full text

MANAGEMENT SCIENCE
Vol. 54, No. 9, September 2008, pp. 1652–1665
issn 0025-1909 eissn 1526-5501 08 5409 1652
informs®
doi 10.1287/mnsc.1080.0877
© 2008 INFORMS
Look Before You Leap: Market Opportunity
Identification in Emerging Technology Firms
Marc Gruber
College of Management of Technology, Ecole Polytechnique Fédérale de Lausanne,
CH-1015 Lausanne, Switzerland, marc.gruber@epfl.ch
Ian C. MacMillan, James D. Thompson
The Wharton School, University of Pennsylvania, Philadelphia, Pennsylvania 19104
{macmilli@wharton.upenn.edu, jamestho@wharton.upenn.edu}
Entrepreneurs play a fundamental role in bringing new technologies to market. Because technologies are
often configurable to serve a variety of different markets, it is possible for entrepreneurs to identify multiple
market opportunities prior to the first market entry of their emerging firms, and if they elect to do so, to therefore
have a choice of which market to enter first. The empirical results presented in this paper offer three new insights
regarding this important early-stage choice in new firm creation. First, they reveal that serial entrepreneurs have
learned through prior start-up experience to generate a “choice set” of alternative market opportunities before
deciding which one to pursue in their new firm creation. Second, the analysis indicates that entrepreneurs who
identify a “choice set” of market opportunities prior to first entry derive performance benefits by doing so.
Third, the positive relationship between the number of market opportunities identified prior to first entry and
new firm performance is nonlinear and subject to decreasing marginal return. The research literature has yet to
acknowledge the notion of multiple opportunity identification prior to entry, and the related idea of selecting
the most favorable market opportunity for the creation of a new technology firm.
Key words: market opportunities; serial entrepreneurs; technological commercialization; new firm creation
History : Accepted by Scott Shane, technological innovation, product development, and entrepreneurship;
received June 3, 2007. This paper was with the authors 1 month and 1 week for 1 revision. Published online
in Articles in Advance July 10, 2008.
1. Introduction
A growing number of studies highlight the fact
that new technology firms are an important source
of innovation and wealth creation (Shane 2004).
Before entrepreneurs can exploit the value inherent in
their technological competences, however, they need
to identify at least one market domain in which
their technologies meet customer demand. Notably,
because technological competences may create bene-
fits for end users in multiple market domains (Penrose
1959, Prahalad and Hamel 1990, Jolly 1997, Danneels
2007), by identifying more than one market opportu-
nity, entrepreneurs might be able to select the most
favorable market conditions for new firm creation.
However, due to limited prior knowledge of mar-
kets in which a technological competence may be
valued, the identification of multiple market opportu-
nities prior to the first entry appears to be fairly un-
common among nascent entrepreneurs. Shane (2000)
shows that in a set of eight entrepreneurs who
sought to commercialize a technology from MIT,
none identified more than one market opportunity,
although at least eight different markets existed.
Intriguingly, these market opportunities offered highly
diverging prospects for value creation. Hence, for an
entrepreneur seeking to establish her new firm as a
prospering economic entity, the failure to identify a
major market opportunity prior to first entry may be
particularly problematic.
Research that could provide further insights on the
technology-to-market linking problem in new firm
creation is surprisingly scant (Helfat and Lieberman
2002), given that this problem is of considerable theo-
retical interest for the entrepreneurship, strategy, and
organizational literatures (Dougherty 1992) and is also
of high practical relevance (Jolly 1997, Shane 2004).
Beyond the value creation aspect, the choice of which
market to enter is one of the most profound organiza-
tional decisions entrepreneurs are faced with, because
the nature of the market has strong imprinting effects
on a new firm’s identity, the capabilities and assets
it needs to build, and its organizational structure
(Boeker 1989).
Drawing from the literatures on learning and inno-
vation and resource-based theory, the purpose of
this paper is to advance theoretical understand-
ing of the technology-to-market linking problem in
new firm creation. We address two important yet
unanswered questions. First, against the backdrop
of recent studies that uncover important effects of
1652

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1653
prior entrepreneurial experience on the process of
opportunity identification (McGrath and MacMillan
2000, Baron and Ensley 2006), we seek to understand
whether serial entrepreneurs (i.e., individuals who are
proficient in new firm creation) have developed par-
ticular approaches to the technology-to-market link-
ing problem and the associated search for market
opportunities. Specifically, we examine whether serial
entrepreneurs identify more market opportunities for
their technologies prior to first entry—i.e., construct
a larger choice set—than novice entrepreneurs. Sec-
ond, there is a major gap in our understanding
of whether entrepreneurs may indeed benefit from
the identification of multiple market opportunities.
Whereas the MIT example shows that market oppor-
tunities stemming from the same technology may
have substantially different value creation character-
istics (Shane 2000), there is also research that suggests
that entrepreneurs may not be able to select effec-
tively between opportunities or may not be able to
exploit opportunities that are more distant from their
current knowledge base. We thus investigate whether
the identification of multiple market opportunities
prior to first market entry leads to superior perfor-
mance outcomes in new firm creation. We address
these research questions using an original data set of
83 venture capital (VC)-backed technology ventures.
The scarcity of studies on the technology-to-market
linking problem in new firm creation conceivably
warrants any addition to the literature. We believe
that several insights of the present study make a con-
tribution. Most important, our findings reveal that a
key element of commercialization learned by technol-
ogy entrepreneurs through prior start-up experience
is to generate a choice set of alternative market oppor-
tunities before deciding which opportunity to pur-
sue first. Furthermore, the analysis indicates that
entrepreneurs can derive performance benefits from
the identification of a set of market opportunities
prior to first entry, and that a positive yet nonlin-
ear relationship exists between the number of mar-
ket opportunities identified (i.e., the size of the choice
set) and firm performance. We proceed with a discus-
sion of the theoretical background and develop our
hypotheses in §3. We describe the estimation method-
ology and the data in §4 and present the empirical
results in §5. Section 6 concludes the paper.
2. Theoretical Background: Linking
New Technologies with
Market Opportunities
In its most basic form, technology-to-market link-
ing can be seen as the combination of technological
knowledge with information on market demand, i.e.,
end-user domains where the technology, as embodied
in a product, can be meaningfully employed and can
create benefits for its users (Dougherty 1992, Danneels
2002). To understand the technology-to-market link-
ing problem in new firm creation, it is necessary to
delineate the components of this problem.
2.1. Technological Competences and
Their Fungibility
Following Mitchell (1992), technological competence
typically stems from a combination of tangible and
intangible technically related resources. The idea that
technological competences are fungible and can create
benefits for end users in multiple market domains is
firmly anchored in the resource-based view in strate-
gic management. As a case in point, Penrose (1959,
p. 25) observed that “(e)xactly the same resource
when used for different purposes or in different ways
and in combination with different types or amounts of
other resources provides different service ().” Sub-
sequently, many authors have noted that technolog-
ical competences lie upstream from the end product
and transcend any particular product (e.g., Prahalad
and Hamel 1990, Danneels 2007). Because a firm’s
technological competence is distinct from the prod-
ucts in which it is embodied, it may be utilized in a
range of applications, and thus leveraged across dif-
ferent market domains.1
At this point it is important to note that the range
of market opportunities emerging from a new technol-
ogy and thus the generality of a technological com-
petence are rarely clear from the outset, because they
depend on people’s (ingenious) efforts in technologi-
cal competence building and market opportunity iden-
tification (Penrose 1959). First, from a technology per-
spective, fungibility depends on how far and in what
direction the knowledge components embodied in a
new technology are recombined with other bits of tech-
nological knowledge (Fleming and Sorenson 2004).
Whereas some elements of a technological competence
are specific to a particular application area, others are
of a more general nature and have implications for
products in a great array of markets. Hence, although
research and development efforts may be focused on
particular applications, emerging technologies may be
untangled, altered, and integrated with other knowl-
edge bases in order to be transformed into a more
general, or more specific, type of technological compe-
tence (Galunic and Rodan 1998, Danneels 2007). Sec-
ond, from a market perspective, fungibility depends
on the identification of different markets in which the
same type of technological functionality is valued, and
thus on the extensiveness of the search process and
1The logic behind this argument corresponds to the rationale for
related diversification (e.g., Miller 2006), which involves expansion
into different markets based on shared organizational competences.

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
1654 Management Science 54(9), pp. 1652–1665, © 2008 INFORMS
the ingenuity that founders apply in their explorations.
To identify promising market opportunities, a complex
array of insights into application domains and cus-
tomer needs must be gathered (Dougherty 1992).
2.2. Acquiring Knowledge on
Market Opportunities
The acquisition of knowledge on market opportuni-
ties can be conceptualized as an organizational search
problem. In the organizational learning literature,
search activities are considered instrumental for firms’
adaptation to environmental change and the iden-
tification of opportunities to enhance performance.
Building on March and Simon (1958), local search is
defined as the behavior of a firm in seeking solu-
tions in the immediate neighborhood of its existing
stock of knowledge. Because local search is associated
with less-risky search activities, it is the most com-
monly used algorithm in technological search (Stuart
and Podolny 1996). In terms of market search, this
suggests that entrepreneurs tend to identify a mar-
ket opportunity either known to them in the past, or
closely related to their existing stock of prior knowl-
edge (Shane 2000). At the other end of the spec-
trum, distant search (also referred to as exploration)
is defined as the behavior of people attempting to
build new knowledge bases (March 1991). To con-
duct a distant search, entrepreneurs need to engage
in some form of boundary spanning and bridge dis-
parate knowledge domains (Miller et al. 2007).
The process of market search and its outcome
depend on whether one assumes rational or bound-
edly rational behavior by founders. In a classic model
of rational decision making, founders are assumed
to explore the complete landscape and evaluate the
whole set of alternatives for linking their technolog-
ical resources with market opportunities (Janis and
Mann 1977). Behavioral decision theorists stress, how-
ever, that people’s search activities are strongly influ-
enced by cognitive and social phenomena (Cyert and
March 1963). Specifically, individuals have limited
information-processing capacity, which can readily be
exceeded when they perform complex tasks such as
organizational searches. This deficiency is augmented
by the uncertainty, resource constraints, and time
pressures experienced in new firm creation. Realisti-
cally, one can thus expect individuals to engage in
a boundedly rational search by employing simpli-
fied representations of search landscapes and decision
heuristics (Gavetti and Levinthal 2000).
If boundedly rational founders cannot readily iden-
tify all solutions, the question becomes how many
do they identify before they terminate their search
for market opportunities? Behavioral decision theory
suggests that actors search for alternatives until they
identify one that is good enough to meet an initial
set of requirements, i.e., they satisfice (Simon 1982).
Search is seen to commence when the newly identi-
fied alternative does not meet the aspiration levels of
founders, and to stop when a satisfying alternative
has been identified. In new firm creation such differ-
ences in aspiration levels may be most evident when
comparing founders who seek to create “lifestyle ven-
tures” with those attempting to create high-growth,
VC-backed businesses. What is considered a satisfac-
tory alternative is typically a function of an actor’s
prior experience and her observation of other referent
organizations (Greve 2003).
3. Hypotheses
The preceding discussion indicates that the initial
market choice is one of the most profound decisions
in the early life of firms and that some heterogene-
ity can be expected in how extensively entrepreneurs
engage in a market opportunity search prior to
first entry. Extending this discussion, our hypotheses
study two critical yet hitherto unaddressed questions.
First, against a backdrop of recent research, which has
uncovered that serial entrepreneurs possess special
insights on opportunity identification (McGrath and
MacMillan 2000, Baron and Ensley 2006), we seek to
understand whether they also possess special insights
on the technology-to-market linking problem in new
firm creation. We ask whether they identify more
market opportunities for their technological compe-
tences prior to first entry—i.e., construct a larger
choice set—than novice entrepreneurs. Second, theory
offers strongly conflicting arguments regarding the
core question of whether entrepreneurs may indeed
benefit from the identification of multiple opportu-
nities prior to first entry (cf. March 1991, Peteraf
and Bergen 2003). Our second set of hypotheses thus
investigates the relationship between preentry market
opportunity search and postentry performance.
3.1. Prior Entrepreneurial Experience and
Market Opportunity Search
Individuals who create new firms are equipped with
a stock of knowledge that they can apply in the
process. Because new firms in technology-intensive
industries are typically founded by teams, it is the
preexisting knowledge of the founding team that
affects the knowledge available to the firm, the abil-
ity of the team to access and use the knowledge,
its information-gathering and information-processing
behavior, and the number and variety of solutions
that will be generated (Pelled et al. 1999). One par-
ticularly important type of preexisting knowledge is
the knowledge that has been acquired through prior
entrepreneurial experience, because repeat founders
can draw on high levels of task-specific knowl-
edge and may have obtained special insights on the

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1655
entrepreneurial process. Because prior entrepreneurial
experience has many tacit components that are
learned by doing, it provides a particular type of
knowledge that cannot be acquired easily through
other types of learning (Delmar and Shane 2006). In
this vein, expert information-processing theory sug-
gests that people develop refined and complex cog-
nitive structures (schemata) as they gain experience
in a particular area. These structures may help repeat
founders in processing new information and in unify-
ing disparate sets of information, and also in arriving
at qualitatively more sophisticated judgments (Gagné
and Glaser 1987).
In terms of opportunity identification, these argu-
ments suggest that serial founders could have devel-
oped specific insights on the process of opportunity
identification and have refined cognitive structures in
place that assist in market opportunity search. Two
recent studies support this assumption. First, Baron
and Ensley (2006) find that experienced founders
have acquired richer and more refined cognitive rep-
resentations of business opportunities than novices
(opportunity prototypes), which helps them pursue
opportunities most likely to yield positive financial
outcomes. Whereas novice entrepreneurs emphasize
evaluation criteria such as the novelty of the idea,
the superiority of the product or service, and the
potential to change the industry, repeat entrepreneurs
look for business opportunities that will quickly gen-
erate positive cash flow, have a manageable level
of risk, and solve a customer’s problem. Second,
field research by McGrath and MacMillan (2000) indi-
cates that experienced founders have developed a
distinct entrepreneurial mindset that provides these
individuals with rich cognitive frameworks for iden-
tifying and pursuing business opportunities. They
find that entrepreneurs who have repeatedly created
successful new firms employ so-called “opportunity
registers;” i.e., they identify and keep track of multi-
ple business opportunities before they decide which
one is the best to pursue. Although McGrath and
MacMillan’s observations concern business opportu-
nities as such, the underlying logic may also apply
in the context of market opportunities for technolo-
gies: Serial entrepreneurs seek out a larger num-
ber of market opportunities prior to first entry, so
that they have a larger choice set at hand before
deciding which market opportunity to pursue with
their emerging technology firms.2As indicated above,
2As an illustration of the multiple market opportunities that can
arise from a technological resource, consider the example of a
young European plastic manufacturer. This firm developed a new
form of plastic that remains water resistant while in use, but can be
dissolved when a reagent is added to the water. It identified mul-
tiple markets for its innovative plastic, with applications ranging
from diapers to humidity sensors to agricultural films (Jolly 1997).
serial entrepreneurs may not only possess superior
abilities in market opportunity identification, but may
also have a higher awareness of the importance of
market conditions for successful new firm creation
and thus may be more willing to search for markets.
Hypothesis 1 (H1). Founding teams that have mem-
bers with prior entrepreneurial experience will consider a
larger number of market opportunities prior to first market
entry than teams without such experience.
3.2. Preentry Market Opportunity Search and
Postentry Performance
Given the uncertainties associated with market oppor-
tunity identification and evaluation, the fleeting
nature of some opportunities, and a variety of other
factors that impact search and firm creation processes,
it is unclear whether performance benefits can be
derived from the identification of multiple oppor-
tunities. Several theoretical arguments suggest that
the identification of multiple market opportunities is
an important precursor to achieving superior perfor-
mance outcomes in new firm creation.
First, it is one of the premises of classical decision-
making theory that the availability of alternative solu-
tions will enhance organizational problem-solving
outcomes (Janis and Mann 1977). This notion is
echoed by behavioral decision theorists (March and
Simon 1958) and also in studies on human creativity
suggesting that people who can brainstorm several
solutions to a problem have a higher likelihood of
finding the most promising solution (Osborn 1957).
Second, market characteristics are of great impor-
tance in entrepreneurship because they have a con-
siderable impact on the rent-earning potential of firm
resources (Peteraf and Bergen 2003). Market oppor-
tunities usually vary along key dimensions such as
market size, lifecycle stage, demand uncertainty, entry
barriers, and competitive rivalry. For example, the
eight markets discussed in Shane (2000) had pro-
jected market sizes that ranged between 10 million
and several billion USD. Thus, moving beyond the
local solution by increasing the search space can lead
to highly novel recombinations of knowledge, poten-
tially more valuable opportunities, and a greater set
of opportunities from which to choose. Research on
established firms supports this reasoning by suggest-
ing that whereas local search can lead to the identi-
fication of local performance optima, above-average
firm performance relies more heavily on a firm’s abil-
ity to identify a global optimum through a distant
search (e.g., Rosenkopf and Nerkar 2001).
Third, the market choice defines a core element of a
new firm and has a strong imprinting effect, because
it forms the basis for other key strategic decisions
such as the configuration of the internal and exter-
nal value chain (Abell 1980, Geroski 1998, Danneels

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
1656 Management Science 54(9), pp. 1652–1665, © 2008 INFORMS
2003). New firms can get stuck with their initial (infe-
rior) choice, because they may not be in a position to
afford the adjustment costs incurred by changing the
target market.
Fourth, although the exploitation of more distant
market opportunities requires an adequate knowl-
edge base and access to appropriate core and com-
plementary resources, some managerial practices can
support entrepreneurs in pursuing distant configura-
tions. For example, they may acquire distant knowl-
edge by hiring new team members, or by creating
alliances (Roy 2005).
Against this backdrop, we argue that it is impor-
tant for successful firm creation that technology
entrepreneurs have a choice of target markets prior
to first entry. We thus postulate the following
hypothesis:
Hypothesis 2A (H2A). Technology start-ups that
consider more than one market opportunity for a tech-
nological competence prior to the first market entry will
be more successful than those that consider just a single
market opportunity.
Extending further, the question arises whether
founders who search for ever-more market oppor-
tunities can expect to benefit from constructing a
larger choice set of such opportunities prior to first
entry. To construct a larger choice set, founders need
to broaden the scope of their search and explore
more distant regions of the search landscape (March
1991). As discussed, distant searches allow a com-
bination of existing knowledge with new, far-flung
knowledge elements. Although research indicates that
the integration of distant knowledge elements may
not necessarily lead to a viable combination, it also
suggests that combinations involving distant knowl-
edge can lead to highly valuable and potentially
path-breaking solutions (Fleming and Sorenson 2004,
Nerkar and Roberts 2004). Thus, chances are that
founders may indeed identify highly attractive mar-
ket domains when they explore more distant regions
of the market search landscape.
However, the construction of a larger choice set
is also associated with additional search costs and
mounting challenges in opportunity evaluation and
exploitation. Specifically, founders will need to devote
increasing amounts of monetary resources, time, and
energy to extend their exploration of the market
landscape (March 1991, Gavetti and Levinthal 2000).3
Thus, when entrepreneurs extend their market oppor-
tunity search they may be able to do so with only few
3Note that some of the costs mentioned in the following discussion
could be of differential importance for novice and for experienced
entrepreneurs. For example, the latter may have a social network in
place that can provide different kinds of resources and capabilities
for new firm creation.
resources and may thus fail to identify major oppor-
tunities. Those who engage in an extended search
may also have fewer resources available for oppor-
tunity exploitation (Gifford 1992, 1998), and may
face an important opportunity cost, or cost of delay
(Radner 1996), since the opportunities identified early
on could be of a fleeting nature.
Further, decision processes will become more ardu-
ous and costly, and the limited information-process-
ing capacity of entrepreneurs will be increasingly
strained when more information has to be collected
and more market alternatives have to be evaluated
(Simon 1982). In this vein, research on cognitive
abilities indicates that the most complex set of inter-
relationships an individual can process in working
memory is a three-way interaction (Halford et al.
1994). Thus, beyond some number of market oppor-
tunities in a choice set, founders may become over-
whelmed by the complexity of the evaluation task
and may not be able to evaluate the causal relation-
ships between potential alternative actions and pos-
sible outcomes (Radner 1996).4In the extreme, they
may even become dysfunctional as they suffer from
cognitive distress or paralysis by analysis (Peters and
Waterman 1982).
It can also be expected that the costs of exploit-
ing more distant market opportunities are higher than
for closely related opportunities. The more distant
the identified market opportunities are, the higher
the costs of acquiring and integrating the knowledge
required to exploit them will be (Katila and Ahuja
2002). Similarly, acquiring or developing the required
resources may be too costly or may take too long to
be practical (Dierickx and Cool 1989).
In sum, these arguments suggest that although an
extended search for market opportunities can lead
to a larger choice set with potentially more valu-
able options, there are also important costs and major
challenges associated with a more extensive search
and evaluation process. Given the uncertainty of an
extended search and these costs and challenges, we
believe that, overall, there are decreasing marginal
returns in constructing a larger choice set.5It should
be emphasized, however, that the specific nature of
4In this vein, the uncertainty associated with more distant opportu-
nities may make it difficult, or even impossible, to arrive at proper
evaluations (Knight 1921).
5There could also be economies of scale in opportunity search
activities. For example, entrepreneurs engaging in a more exten-
sive search may purchase third-party databases to acquire infor-
mation on potential application areas. Because of the possibility
of increased search efficiency with such inputs, a decrease in the
average cost of identifying additional market opportunities may
be observed. However, following the arguments presented above,
we believe that there are increasing challenges and costs associated
with performing a distant search for market opportunities.

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1657
this relationship is ultimately an empirical question—
one that we investigate with this hypothesis:
Hypothesis 2B (H2B). The performance benefit of
considering more than one market opportunity is subject
to decreasing marginal returns.
4. Research Design
4.1. Sample
To examine the research questions outlined previously,
we required data on the founding team, preentry mar-
ket opportunity identification, firm performance, and
other organizational characteristics. No public data set
offers such information, so we collected data through a
self-administered survey of new firms backed by ven-
ture capital. With this sample of new firms, we could
be confident that we were studying performance-
oriented new firms with high growth aspirations.
Data for this study comes from two sources:
(1) a questionnaire addressed to founders of VC-
backed companies in Germany, and (2) a separate
data set containing performance data on a subset
of the surveyed firms. Survey data were collected
in the summer of 2003 using a standardized Web-
based questionnaire. Access to the sample of 348
VC-backed firms was provided through collabora-
tion with a professional services firm specializing
in Web-based solutions for the VC industry. Follow-
ing a comprehensive pilot study that included inter-
views with more than two dozen entrepreneurs, we
developed the survey instrument and pretested it
with 10 entrepreneurs, 4 venture capitalists, and 9
academics. Prior research considers founders highly
knowledgeable and valid information sources, so we
addressed the questionnaires directly to the firms’
founders, following a key-informant approach (Huber
and Power 1985, Glick et al. 1990). The questionnaire
was online for 12 weeks. A telephone follow-up was
conducted after Week 6. We received questionnaires
from 142 VC-backed firms, yielding a response rate
of 40.8%. Respondent firms had a median founding
year of 1999. We conducted an analysis for nonre-
sponse bias by comparing early and late respondents,
with the latter serving as a proxy group for nonre-
sponding firms (Armstrong and Overton 1977). No
indication of nonresponse bias was found.
We merged the survey data set with a second data
set obtained from our collaboration partner and con-
taining information on firm performance. Specifically,
this data set provided monthly revenue data, which
we totaled to obtain the yearly revenue figures used
in our analysis. However, because not all ventures
obtained venture capital prior to their initial mar-
ket entry (see statistics provided further below) and
because some VC-backed ventures joined the report-
ing platform of the collaboration partner only after
they had performed their market entry, the second
data set provided revenue data for the first and sec-
ond year after market entry only on a subset of ven-
tures represented in the first data set. The merged
data set contained full observations from a total of 83
VC-backed ventures. We analyzed whether any bias
was introduced because of the reduced sample size
by comparing firms in our merged data set with firms
that were not part of the merged data set. No indica-
tion of such a bias was found.
4.2. Definition and Measurement of Variables
4.2.1. Dependent Variables (Market Opportu-
nity Count Models/Performance Models). Number
of Alternative Market Opportunities. This variable re-
cords the number of alternative market opportunities
entrepreneurs considered for their technological com-
petence prior to the first market entry. In a multi-
stage question, respondents were asked to indicate
whether, prior to first market entry, they had con-
sidered commercializing their technological resources
(respectively, competences or know-how) in mar-
ket domains completely different from the market
entered. Respondents who had indicated that they
had considered alternative markets were then asked
to report the number of alternatives considered.
Sales Revenues. Annual sales revenues (in euros)
achieved by the new venture in Year 1 and in Year 2
after the initial market entry were used as dependent
variables. Following Fox (1991), we utilized logarith-
mic transformations (natural logarithms) for the pos-
itively skewed sales variables.6
4.2.2. Covariates. Prior Entrepreneurial Experience.
Respondents reported the number of members of
the founding team who had previously started a
new firm. We created a dummy variable to capture
whether the founding team possessed prior entrepre-
neurial experience (1) or not (0).
Management Experience/Technological Experience/Mar-
keting Experience. Following Wiersema and Bantel
(1992, p. 95), we used the average level of a given trait
in a team to represent the group’s overall character-
ization. Respondents rated the levels of management
experience, technological experience, and marketing
experience that the team possessed (on a five-point
Likert-type scale, “very low” to “very high”).
6There is no consensus as to what constitutes entrepreneurial
success (Delmar and Shane 2006). Entrepreneurs have differing
objectives for starting new firms (e.g., “lifestyle ventures” versus
“gazelles”), and objectives also vary in importance at different
stages of firm creation. Our study analyzes a sample of VC-backed
ventures, i.e., new firms that are highly growth oriented. The
achievement of sales revenues once a new firm has entered the
market is important, because it shows financiers and other stake-
holders that a firm’s vision of the market is real (“proof of market”)
and that it can deliver on its vision.

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
1658 Management Science 54(9), pp. 1652–1665, © 2008 INFORMS
Total Time Spent on Planning and Organizing the Ven-
ture. Searching for additional opportunities involves
both direct and indirect opportunity costs. We thus
controlled for the total time spent on planning and
organizing the new firm. The measurement of plan-
ning duration utilized in this study follows prior
studies (Brüderl et al. 1996) and is similar to mea-
sures in strategy research (Brews and Hunt 1999). We
measured how many months it took from the start of
active preparation of the new firm to its market entry.
VC Seed Funding Prior to Market Entry. Extant
research suggests that the availability of resources
will influence founding team behavior (e.g., Helfat
and Lieberman 2002). Access to financial resources
will likely impact market opportunity search, because
entrepreneurs with a resource buffer should be less
pressed for a speedy market entry and can spend
more on search activities and on reconfiguring their
knowledge/resource base to perform long jumps on
the search landscape. Although all firms in our sam-
ple were backed by venture capital, not all of them
obtained it during the seed stage prior to market
entry. A dummy thus indicates whether the emerging
firm acquired seed funding (1) or not (0).
Total Costs in Year 1/Year 2. The total costs incurred
by the new firm in the first and in the second year
after market entry are used as controls in the first-
and second-year performance regressions, respec-
tively. Cost data were obtained from the database of
the collaboration partner.
Breadth of Market Entry. New firms vary in the
breadth of their product offerings when entering mar-
kets, so we created a dummy variable that indicates
whether the new firm entered with multiple prod-
ucts (0) or with a single product (1).
Founding Team Size. Because team size is an impor-
tant indicator of the human capital available in a new
firm, market opportunity search may be influenced by
founding team size. Following Wiersema and Bantel
(1992), we used a count of the individuals in the ini-
tial team (prior to first entry).
We also used several controls pertaining to the tech-
nologies employed by the emerging firms:
Technological Fields. The resource requirements to
develop new technologies and other factors vary
across technological fields. To parcel out such varia-
tion, we included dummies to control for the tech-
nological fields represented in our study: Multimedia
and Communication (11% of the sample), Inter-
net (12%), Electronics (6%), Process Engineering
(13%), Software (19%), New Materials (5%), Measur-
ing/Regulation Technology and Handling Systems
(20%), and Others (14%).
Generality of Technology. Whereas the dummies for
technological fields are “catch-all” measures to control
for variations across technologies, we also employed
a more specific measure to parcel out the generality
of employed technological competences in our mar-
ket opportunity count models. The ex ante determi-
nation of the generality of an emerging technological
competence is, however, not only a major theoretical
problem (Bresnahan and Trajtenberg 1995, cf. §2.1),
but also poses a difficult empirical challenge. We took
advantage of the information contained in our data
set and utilized the full number of observations on
market opportunity identification (n=133) to denote
technological fields in which entrepreneurs identify
additional market opportunities more frequently. The
derived index captures the share of new firms in
a particular technological field that considered more
than one market opportunity.7
Self-Developed Technology Licensed-In Technology.
Using a five-point Likert-type scale, respondents
reported the degree to which the technology they
sought to commercialize had been licensed in (1) or
developed internally (5).
4.3. Methods
We used two statistical methods to test our
hypotheses. Our first dependent variable—number of
alternative market opportunities—takes on only nonneg-
ative integer values and consists of zeros and ones
to a significant degree. An ordinary linear regression
is not applicable in this context, because it relies on
the normality of the dependent variable (Wooldridge
2002, Greene 2003). Poisson regression can be used
to model count variables, but the negative binomial
regression model is preferred when the assumption
of the equality of the conditional mean and variance
functions is violated (Greene 2003). Because a like-
lihood ratio test indicated overdispersion, we used
the negative binomial model, which generalizes the
Poisson model by introducing an individual unob-
served disturbance term that allows the mean and
variance to vary (Hausman et al. 1984). The negative
binomial model (cf. Greene 2003) takes the following
form: ln i=xi+i(iequals the mean and vari-
ance of yi,xiis the vector of regressors, and exp
is the gamma-distributed error term). For our second
set of dependent variables, sales revenues in Year 1
and Year 2 after market entry, we employ an ordinary
least-squares estimator (Wooldridge 2002).
Recent publications (Hamilton and Nickerson 2003)
suggest that endogeneity might pose a problem in
empirical management research, because managers
(entrepreneurs) make decisions based on expectations
7Index values: Multimedia and Communication (0.27), Internet
(0.17), Electronics (0.33), Process Engineering (0.17), Software (0.25),
New Materials (0.17), Measuring/Regulation Technology and Han-
dling Systems (0.26), and Others (0.30). In unreported robustness
checks, we find that results are consistent when employing this
measure in lieu of the technological field dummies.

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1659
of how their choice impacts future firm performance.
In particular, self-selection on unobservable or hard-
to-measure factors may simultaneously influence mar-
ket search and performance. To address potential
problems arising from endogeneity in the market
opportunity count models, we followed Heckman’s
(1979) method to correct for self-selection, because
2SLS and 3SLS models are not applicable in the
context of negative binomial regressions (Wooldridge
2002). We first estimated a reduced-form probit model
capturing the decision to consider additional market
opportunities. Second, we estimated market opportu-
nity counts as a function of identified variables and
corrected for possible self-selection bias by including
the inverse Mills ratio, i.e., an index that was gener-
ated from the probit estimates. To address potential
problems arising from endogeneity in the perfor-
mance models, we performed instrumental variables
regression (by first predicting the number of alter-
native market opportunities with a set of exogenous
variables in a negative binomial model, then utiliz-
ing the predicted number in our performance models);
we report these results in Table 3 (Models 4 and 8).
Moreover, we include a lagged dependent variable
in the second-year models (Models 5–7), because the
inclusion of this variable offers a straightforward way
to correct for potential biases from unobserved fac-
tors that codetermine market search and performance
(Wooldridge 2002, p. 300; Hamilton and Nickerson
2003). We will thus be able to obtain accurate estima-
tions of the ceteris paribus effect of market opportu-
nity identification.
5. Empirical Results
5.1. Descriptive Statistics
The descriptive statistics and the correlation matrix
are reported in Table 1. Correlations are relatively low,
indicating that collinearity should not be a concern.
We see that 28% of the firms considered more than one
market opportunity prior to first entry. This evidence
provides an important extension to Shane’s (2000)
qualitative study of the eight MIT entrepreneurs, each
of whom identified just one market opportunity prior
to entry. Notably, we find that firms with multiple
market opportunities can be found in all technolog-
ical fields represented in our study (see values in
Footnote 7). Figure 1 depicts the number of addi-
tional market opportunities that had been considered
prior to first entry. In the subsample of new firms
that considered multiple markets, the majority con-
sidered three additional markets (subsample mean =
368; median =3). A separate calculation shows that
firms that considered multiple opportunities achieved
median revenues in the first and second year after
Table 1 Descriptive Statistics and Correlation Matrix
Variable Mean S.D. 1 2 345678910111213141516
1 Revenues Year 1 (log) 1001 359 100
2 Revenues Year 2 (log) 1172 188 058∗∗∗ 100
3 Alternative market opportunity 028 045 051∗∗∗ 038∗∗∗ 100
4 Number of market 035 062 048∗∗∗ 035∗∗∗ 098∗∗∗ 100
opportunities (log)
5 Total time spent organizing 1506 1084 −011 −032∗∗∗ −012 −010 100
and planning the new firm
6 VC funding prior to market entry 053 050 028∗020†−001 −003 −013 100
7 Breadth of market entry 023 042 −005 000 −015 −016 −007 005 100
8 Self-developed technology 469 068 −010 −006 004 004 013 −006 001 100
9 Generality of technology 025 021 −012 014 −001 000 −010 −005 −005 013 100
10 Total costs in Year 1 1340 093 005 027∗006 006 −037∗∗∗ 019†004 017 025∗100
11 Total costs in Year 2 1338 086 007 037∗∗∗ 005 004 −041∗∗∗ 009 008 016 023∗086∗∗∗ 100
12 Number of firm founders 277 157 021†018†008 008 −002 006 002 005 −010 −002 003 100
13 Prior entrepreneurial experience 042 050 011 025∗018 019†−014 017 005 −008 011 008 008 003 100
14 Management experience 302 106 018†012 027∗028∗−011 012 007 014 024 016 012 007 030∗∗ 100
15 Marketing experience 272 105 −005 012 005 007 −017 003 014 005 046∗∗∗ 018†018†−010 039∗∗∗ 063∗∗∗ 1
16 Technological experience 448 074 −007 −008 −016 −016 003 005 −015 021 −017 006 007 −002 009 −008 −017 1
†p<010, ∗p<005, ∗∗p<001, ∗∗∗p<0001.

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
1660 Management Science 54(9), pp. 1652–1665, © 2008 INFORMS
Figure 1 Market Opportunity Count of Emerging Technology Firms
Prior to First Market Entry
80
10
123 4>4
100
90
70
60
0
72.3%
6.0%
8.4%
2.4%
3.6%
7.3%
Number of additional market opportunities
Percentage of new firms (%)
Note. n=83.
entry of E145,000/E329,000, whereas firms that consid-
ered a single market opportunity achieved revenues
of E33,000/E121,000.
5.2. Market Opportunity Count Models
Results of the negative binomial regression are pre-
sented in Table 2. Model 1 estimates a baseline model
of controls, Model 2 adds the predictor variable prior
entrepreneurial experience, and Models 3 and 4 esti-
mate interaction effects of team variables. The predic-
tor variable and the interactions significantly increase
the explanatory power of Models 2–4, as measured by
twice the difference in the respective log-likelihoods
and compared to a chi-square statistic with degrees
of freedom equal to the number of newly added vari-
ables. Note that we utilize conservative two-sided
tests of significance in Table 2.
Hypothesis 1 predicted that founding teams with
prior entrepreneurial experience would identify a
larger number of market opportunities than teams
lacking this experience. We find support for this
hypothesis in Model 2 of Table 2. Whereas the effects
of prior entrepreneurial experience and management
experience are positive, technological experience
and marketing experience have a negative associ-
ation. Additional support for Hypothesis 1 is pro-
vided in Models 3 and 4, which offer a more
detailed examination of the knowledge available in
founding teams by including interaction terms. We
see that prior entrepreneurial experience positively
moderates the relationships between marketing expe-
rience and market opportunity identification, and
between technological experience and market oppor-
tunity identification. Because interaction effects do not
lend themselves to easy interpretation in nonlinear
estimation models, we provide a pictorial representa-
tion of these interactions in Figures 2a and 2b. Both
graphs show that teams with prior entrepreneurial
experience consider a larger number of market oppor-
tunities prior to first market entry. Notably, these
Table 2 Negative Binomial Regressions: Models of Market
Opportunity Counts
Model 1 Model 2 Model 3 Model 4 Model 5
Coeff. Coeff. Coeff. Coeff. Coeff.
Variable (S.E.) (S.E.) (S.E.) (S.E.) (S.E.)
Prior entrepreneurial # 080†125∗078†080†
experience 044053045045
Management experience 078∗∗ 070∗093∗∗ 065∗073∗
030030033030031
Marketing experience −039 −053†−146∗∗ −046 −057†
028029051030031
Prior entrepr. exp. # # 126∗##
×Marketing exp. 052
Management exp. ####004
×Marketing exp. 018
Technological −054∗∗ −067∗∗ −079∗∗ −086∗∗ −059†
experience 025025028028031
Prior entrepr. exp. # # # 069†#
×Technol. exp. 038
Management exp. ####−013
×Technol. exp. 028
Number of −001 001 −005 −002 −001
firm founders 010010011010011
Self-developed 043 045 037 067 040
technology 043042043046042
VC seed funding prior 065 064 068 082†063
to market entry 047047049050048
Generality of 105 089 058 107 096
technology 082084085087087
Inverse mills ratio 039 040 076†027 041
(selection correction) 042042046044043
Constant −370†−420†−473∗−530∗−399†
218216222238215
Log-likelihood −9237 −9071 −8730 −8904 −9055
Observations 83 83 83 83 83
†p<010, ∗p<005, ∗∗p<001, ∗∗∗p<0001, #not included, two-tailed
tests.
effects do not show up in interactions with manage-
ment experience (Model 5).
5.3. Performance Models
To test the performance hypotheses, we have esti-
mated two sets of regression models (Table 3). The
first set of estimations (Models 1–4) gives results
for performance in the first year after market entry.
Model 1 reports the baseline model in which a num-
ber of control variables, including the technology field
dummies, were introduced. In Model 2 we intro-
duce the dummy variable capturing the consideration
of alternative market opportunities prior to the first
entry. In Model 3 we substitute the dummy variable
for the number of market opportunities considered
in order to assess the functional form of the rela-
tionship between market opportunity identification
and new firm performance. In Model 4 we substitute
the variable number of market opportunities considered
with an instrumental variable derived in a two-stage
estimation procedure (see above). The second set of

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1661
Figure 2a Interaction Effect of Prior Entrepreneurial Experience and
Marketing Experience
0
1
2
3
1.67 2.72 (mean) 3.77
Marketing experience
Additional market opportunities
Without prior
entrepreneurial
experience
With prior
entrepreneurial
experience
Figure 2b Interaction Effect of Prior Entrepreneurial Experience and
Technological Experience
Without prior
entrepreneurial
experience
With prior
entrepreneurial
experience
0
1
2
3
3.74 4.48 (mean)
Additional market opportunities
5
Technological experience
estimations (Models 5–8) replicates this setup using
second-year performance as the dependent variable.
Again, we use two-sided tests of significance.
In H2A we propose that the consideration of more
than one market opportunity prior to first entry has a
beneficial effect on new firm performance. Even after
controlling for differences across technological fields
using the technological field dummies, the estimates
provided in Model 2 reveal that the coefficient of
the dummy variable more than one market opportunity
is positive and significant. Hence, we claim sup-
port for H2A. New firms that considered more than
one market opportunity prior to first entry generated
significantly higher revenues early on, which is an
important achievement in firm creation.
Hypothesis 2B proposes that the performance ben-
efits of considering multiple market opportunities
are subject to decreasing marginal returns. We tested
this hypothesis by substituting the dummy variable
in Model 2 with the continuous variable number
of alternative market opportunities (logarithmic form)
in Model 3 and the predicted continuous variable
(instrument) in Model 4. The positive and significant
coefficients of both variables suggest support for H2B,
with point estimates from the regression estimation
confirming a positive yet decreasing relationship with
performance (we did not find evidence of an inverted-
U-shaped relationship).
Models 5–8 in Table 3 give results for firm perfor-
mance in the second year after market entry. As men-
tioned, a lagged dependent variable (performance in
the first year) was included to account for unobserved
factors (Wooldridge 2002). Although effect sizes are
somewhat smaller, Models 6–8 mirror the findings
from the first set of regressions, and thus provide
additional support for H2A and H2B. Again, the effect
of identifying an increasing number of markets is
nonlinear; point estimates from the regression equa-
tion confirm a positive, yet decreasing, relationship
with performance.
Among the controls, we confirm earlier studies on
entrepreneurial teams by finding a significant posi-
tive relationship between the number of founders and
firm performance (e.g., Eisenhardt and Schoonhoven
1990). We do not find a significant effect of prior
entrepreneurial experience; this result is in line with
earlier studies showing insignificant effects (e.g.,
Westhead and Wright 1998).
5.4. Robustness Tests
We performed several additional analyses to test
the robustness of our results. First, we utilized the
full sample to analyze the robustness of our market
opportunity count models. Results are consistent with
those of the restricted sample. Second, we performed
a probit estimation utilizing the dummy analyzed addi-
tional markets or not as a dependent measure. Results
are consistent with the negative binomial regressions
of market opportunity counts, in both the restricted
and full samples. Third, to analyze the robustness of
our performance models we utilized the combined
sales performance in the first and second year after
market entry as the dependent variable. The analy-
sis produced results (see Model 9 in Table 3) that
are consistent with those discussed earlier. Fourth,
although profits in the first and second year after mar-
ket entry are a highly debatable measure of the early
performance of strongly growth-oriented ventures,
we nonetheless ran our analysis with this dependent
variable. We did not find that the identification of a
single market opportunity, or multiple ones, is signif-
icantly related to early profit outcomes, nor were any
of the other substantive variables. Given the charac-
teristics of VC-backed ventures, we believe that prof-
its could be a meaningful early performance measure

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
1662 Management Science 54(9), pp. 1652–1665, © 2008 INFORMS
Table 3 Performance Models
Log revenues–
Log revenues–Year 1 Log revenues–Year 2 Years1&2
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9
Variable Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.) Coeff. (S.E.)
More than one market # 108∗∗∗ ###058†## #
opportunity (0/1) 028030
Number of alternative market # # 073∗∗ ###049∗#046∗
opportunities (log) 022021018
Predicted number of 227∗∗ ###135†#
alternative market 084077
opportunities (log)
Log revenues Year 1 ####014∗∗ 011∗∗ 012∗∗ ##
003004003
Total time spent organizing/ −002 −001 −001 −002 −002†−003∗∗ −003∗∗ −004∗∗ −004∗∗
planning the new firm 001001001001001001001001001
VC seed funding prior to 065∗061∗065∗059∗006 020 027 040 077∗∗
market entry 029025025027025025023028022
Breadth of market entry −033 −006 −003 −016 −031 −019 −015 −019 −007
032029030030027028026031025
Number of firm founders 028∗∗ 028∗∗ 029∗∗ 019∗014†016∗016∗009 005
010008008009008008007010007
Prior entrepreneurial −033 −046 −048 −045 037 033 025 044 016
experience 037029029030028028026031025
Management experience 015 003 003 013 −005 −017 −016 006 013
019017017017016017015018015
Marketing experience 015 021 020 017 007 014 018 −005 −008
019017017018017017016019015
Technological experience −005 005 010 001 −027 −023 −015 −025 −010
019017018018016017016018015
Total costs in Year 1 009 006 001 011#### #
018016016017
Total costs in Year 2 ####040∗032∗028†037∗#
018015014017
Dummies for technological Included Included Included Included Included Included Included Included Included
fields
Constant 1071∗∗∗ 861∗∗ 924∗∗∗ 787∗∗ 573∗729∗∗ 728∗∗ 783∗∗ 1211∗∗∗
306238245258266233215256099
Observations 83 83 83 83 83 83 83 83 83
R-squared 052 064 063 060 052 055 060 047 049
Notes. Robust regression results. †p<010, ∗p<005, ∗∗p<001, ∗∗∗p<0001, #not included, two-tailed tests.
in more conventional start-ups, and will become a
more meaningful measure in later stages of growth-
oriented ventures.
6. Discussion
We have argued that the market choice is a profound
organizational decision in emerging firms. Against
a background of sparse prior research, this study
has produced several interesting results that have
novel implications for the organizational, strategy,
and entrepreneurship literatures, and for practice.
6.1. Theoretical Implications
Opportunity identification and exploitation are seen
as the core of the entrepreneurial process (Shane and
Venkataraman 2000). Although opportunity identifi-
cation plays a major role in conceptualizations of firm
creation, the notion of multiple opportunity identi-
fication prior to entry has yet to be acknowledged
in the research literature. By showing that multiple
market opportunity identification matters, this study
suggests that the literature needs to better reflect the
difference between identifying a single opportunity
or multiple ones and the related idea of choosing the
most favorable market conditions for new firm cre-
ation (a notion that evolutionary biologists commonly
refer to as “habitat selection”).
Because there are multiple demands placed on the
limited attention of entrepreneurs, the identification
of market opportunities requires that entrepreneurs

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1663
allocate time and attention away from other, perhaps
more routine, activities to search the market opportu-
nity landscape and gather new information (Radner
1996, Gifford 1998). This allocation of attention comes
at an opportunity cost, or cost of delay, in that less
time and energy, or none at all, is available to exploit
the first identified opportunity. However, there is also
an opportunity cost, and as our findings suggest an
important one, in not seeking to identify additional
markets. Thus, researchers may want to develop a
more nuanced understanding of a “speeding products
to market” approach suggested by earlier studies, as
entrepreneurs run the risk of rushing products to infe-
rior markets when they neglect to explore alternative
options.
The findings also add to our knowledge on the
effects of prior entrepreneurial experience, because
it is not only experience in establishing a new firm
but also in opportunity identification that has a key
impact on new firm performance. Our results pro-
vide an important complement to recent research by
Baron and Ensley (2006) that indicates that serial and
novice entrepreneurs apply different criteria when
assessing a business opportunity. We add to this top-
level picture by revealing that serial entrepreneurs
also identify a larger number of market opportuni-
ties than novices do, and thus are in a position to
choose the most promising market. This view also
extends McGrath and MacMillan’s (2000) character-
ization of the entrepreneurial mindset exhibited by
serial entrepreneurs.
Furthermore, we find that teams possessing a
mix of prior entrepreneurial experience and experi-
ence in technology (or marketing) identify a larger
number of opportunities than teams with technol-
ogy (or marketing) experience only.8Although we
do not observe information flows within teams, the
uncovered pattern indicates that serial entrepreneurs
influence the market search behavior of other team
members. Whereas recent research suggests that it
“is not the selection of people that determines the
degree of exploration, but what they are asked to do”
(Taylor and Greve 2006, p. 736), the present findings
8Whereas these effects are not addressed in the present paper, we
will provide a brief discussion of this pattern. We find that techno-
logical experience is negatively related to market search. Following
prior research in entrepreneurship and innovation (cf. Dougherty
1992), it seems that a primary reason for this finding is that peo-
ple in technology tend to show less appreciation for market-related
topics. There may also be multiple reasons for the negative effect of
marketing experience, but it seems likely that individuals with such
a background are well trained to craft and execute sophisticated
marketing plans, and less so in the earlier stage of market opportu-
nity identification. A thorough analysis of how human and social
capital endowments shape market opportunity identification in
technology firms can be found in our companion paper (Gruber
et al. 2007, 2008).
thus indicate that both aspects matter for exploration:
selecting people with the requisite level of knowl-
edge (e.g., high levels of technological experience)
and engaging them in an exploratory search.
Along these lines, our study has additional impli-
cations for the search literature. First, most studies
in this area focus on technology search (Fleming and
Sorenson 2004). We add to this body of work by
investigating search efforts for market opportunities
for technological resources. Second, prior studies have
devoted only scant attention to the key aspect of
alternative generation (Knudsen and Levinthal 2007).
Our study provides evidence of alternative genera-
tion regarding a major organizational decision, and
shows the empirical relationship between the number
of alternatives generated and performance. Finally,
evidence on the value of path-creating search has
been an important desideratum. As the beginnings
of paths are more clearly visible in new firms, our
research setting provides a vantage point from which
to observe path creation. The results suggest a non-
linear value of path-creating market search.
6.2. Implications for Practice
Most generally, our findings indicate that technol-
ogy entrepreneurs should “look before they leap.”
Although market search activities involve some chal-
lenges and costs, the consequences of not identifying
a market that offers more favorable founding con-
ditions can be profound. It is thus unfortunate that
current business-planning handbooks typically fail
to offer guidelines on opportunity identification and
selection, because they assume that an opportunity
has been identified. We are aware of only two books
in this area that address the issue of opportunity iden-
tification and selection (Mullins 2003, McKnight 2004).
As for investors, VCs may suffer from biases in eval-
uating new firm proposals, or may not be able to
detect when such proposals suffer from entrepreneurs’
perceptual biases. Insofar as biases prevent the consid-
eration of alternative market opportunities, we sug-
gest that investors should emphasize an evaluation of
technological cross-application opportunities in their
due diligence.
6.3. Limitations
In interpreting the results of this study, certain lim-
itations must be kept in mind. First, our analysis is
based on a sample of VC-backed firms, an elite cate-
gory. On the one hand, this sample has advantages for
a study of market opportunity search, because these
firms are begun with high professionalism. Because
just 28% of firms in this sample considered more than
one market opportunity, we believe that this fraction
would have been smaller in samples of more con-
ventional new firms, thus making it more difficult to

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
1664 Management Science 54(9), pp. 1652–1665, © 2008 INFORMS
examine the phenomenon. On the other hand, our
results could be biased toward a specific type of new
firm. Second, our study faced common challenges
associated with measuring the generality of techno-
logical competences. Whereas prior research stresses
resource fungibility (Penrose 1959, Jolly 1997), some
technological competences may be more easily adapt-
able to additional markets than others. In an attempt
to parcel out this type of variation, we employed three
different kinds of control variables for the emerg-
ing technologies, and also controlled for the availabil-
ity of financial resources. Although we consider this
approach to be a viable if imperfect solution to a very
complex problem, we also sought to increase confi-
dence in our results by using instrumental variables
regression and by employing a lagged dependent
variable. The results offered consistent support for
our hypotheses. Third, data from privately held tech-
nology ventures—and, in particular, from VC-backed
ventures—is hard to obtain. In the light of our empir-
ical results it seems that such a sample of highly pro-
fessional ventures is required to address the research
questions examined in our study. However, we also
note that the sample used for the empirical analyses is
relatively small. Fourth, our study is based on perfor-
mance data covering the first two years after market
entry, which is a particularly critical period in the life
of new firms. Although we cannot rule out that new
firms that underperformed in the first two years may
become outperformers at some later point, we believe
that early-stage performance is a major achievement
in new firm creation. For example, early-stage perfor-
mance will signal to investors, potential customers,
and other stakeholders that the new firm is on a solid
track to become an established entity. Nonetheless,
future research needs to explore longer term perfor-
mance. Further, given the discussion about appropri-
ate measures of new firm performance, we note that
this study used data on sales revenues as a perfor-
mance measure. Although the achievement of sub-
stantial sales revenues is an important performance
outcome for VC-backed technology firms, future stud-
ies should also investigate the relationship between
market search and other success measures.
6.4. Further Research and Conclusion
The findings from this study suggest several inter-
esting opportunities for future research. Because our
results indicate that repeat entrepreneurs possess spe-
cial insights on technology-market linking, we sug-
gest that future studies should develop a better under-
standing as to why repeat entrepreneurs have learned
to identify multiple market opportunities prior to
deciding which market to enter. For example, we
believe that repeat entrepreneurs have learned not
only that market conditions are of high importance,
but also that technological fungibility opens up oppor-
tunities for resource leveraging. Along these lines,
researchers may also want to study the capability
needed to perform an effective market opportunity
search. As our companion study suggests, the knowl-
edge repertoire of the founding team and its social
network seem to play a major role in the develop-
ment of this capability (Gruber et al. 2007, 2008). Given
the broader importance of market opportunity iden-
tification for firm emergence, firm evolution, and the
exploitation of the value inherent in resources, we
believe that further studies investigating the nuances
of the market search activity are crucial to under-
standing entrepreneurship, resource leveraging, deci-
sions on organizational scope, growth, and related
phenomena.
Acknowledgments
The authors thank the Fritz Thyssen Foundation, Bonn,
and the Odeon Center of Entrepreneurship, University of
Munich, for their financial support, and Jörn Densing and
Guido Schenk for their support in data collection. This work
benefited from discussions with Pauline Abernathy, Laura
Cardinal, Erwin Danneels, Dietmar Harhoff, Karin Hoisl,
John Mullins, Sonali Shah, Chris Tucci, and Anu Wadhwa.
The authors are also grateful for the insightful feedback
from three anonymous reviewers and the editors.
References
Abell, D. F. 1980. Defining the Business: Starting Point of Strategic
Planning. Prentice Hall, Englewood Cliffs, NJ.
Armstrong, J. S., T. S. Overton. 1977. Estimating nonresponse bias
in mail surveys. J. Marketing Res. 24 396–402.
Baron, R. A., M. D. Ensley. 2006. Opportunity recognition as
the detection of meaningful patterns. Management Sci. 52
1331–1344.
Boeker, W. 1989. Strategic change: The effects of founding and his-
tory. Acad. Management J. 32 489–515.
Bresnahan, T. F., M. Trajtenberg. 1995. General purpose technolo-
gies: “Engines of growth?” J. Econometrics 65 83–108.
Brews, P. J., M. R. Hunt. 1999. Learning to plan and planning to
learn: Resolving the planning school/learning school debate.
Strategic Management J. 20 889–913.
Brüderl, J., P. Preisendörfer, R. Ziegler. 1996. Der Erfolg neugegrün-
deter Betriebe. Duncker and Humblot, Berlin.
Cyert, R. M., J. G. March. 1963. A Behavioral Theory of the Firm.
Prentice Hall, Englewood Cliffs, NJ.
Danneels, E. 2002. The dynamics of product innovation and firm
competences. Strategic Management J. 23 1095–1121.
Danneels, E. 2003. Tight-loose coupling with customers: The enact-
ment of customer orientation. Strategic Management J. 24
559–576.
Danneels, E. 2007. The process of technological competence lever-
aging. Strategic Management J. 28 511–533.
Delmar, F., S. Shane. 2006. Does experience matter? The effect of
founding team experience on the survival and sales of newly
founded ventures. Strategic Organ. 4215–247.
Dierickx, I., K. Cool. 1989. Asset stock accumulation and sus-
tainability of competitive advantage. Management Sci. 35
1504–1511.
Dougherty, D. 1992. Interpretive barriers to successful product
innovation in large firms. Organ. Sci. 3179–202.

Gruber, MacMillan, and Thompson: Market Opportunity Identification in Emerging Technology Firms
Management Science 54(9), pp. 1652–1665, © 2008 INFORMS 1665
Eisenhardt, K. M., C. B. Schoonhoven. 1990. Organizational growth:
Linking founding team, strategy, environment, and growth
among U.S. semiconductor ventures, 1978–1988. Admin. Sci.
Quart. 35 504–529.
Fleming, L., O. Sorenson. 2004. Science as a map in technological
search. Strategic Management J. 25 909–928.
Fox, J. 1991. Regression Diagnostics: An Introduction. Sage, Newbury
Park, CA.
Gagné, R. M., R. Glaser. 1987. Foundations in learning research.
R. M. Gagné, ed. Instructional Technology: Foundations.
Lawrence Erlbaum, Hillsdale, NJ, 49–84.
Galunic, D. C., S. Rodan. 1998. Resource recombinations in the
firm: Knowledge structures and the potential for Schumpete-
rian innovation. Strategic Management J. 19 1193–1201.
Gavetti, G., D. Levinthal. 2000. Looking forward and looking back-
ward: Cognitive and experimental search. Admin. Sci. Quart. 45
113–137.
Geroski, P. A. 1998. Thinking creatively about markets. Internat. J.
Indust. Organ. 16 677–695.
Gifford, S. 1992. Allocation of entrepreneurial attention. J. Econom.
Behav. Organ. 19 265–284.
Gifford, S. 1998. Limited entrepreneurial attention and economic
development. Small Bus. Econom. 10 17–30.
Glick, W. H., G. P. Huber, C. C. Miller, D. H. Doty, K. M. Sutcliffe.
1990. Studying changes in organizational design and effective-
ness: Retrospective event histories and periodic assessments.
Organ. Sci. 1293–312.
Greene, W. 2003. Econometric Analysis. Prentice Hall, Upper Saddle
River, NJ.
Greve, H. 2003. Organizational Learning from Performance Feedback.
Cambridge University Press, Cambridge, UK.
Gruber, M., I. C. MacMillan, J. D. Thompson. 2007. A tale of four
genes: How human capital endowments shape the identifica-
tion of subjective market opportunities in technology start-ups.
Working paper CDM-EPFL & Sol C. Snider Entrepreneurial
Research Center, Wharton School, University of Pennsylvania,
Philadelphia.
Gruber, M., I. C. MacMillan, J. D. Thompson. 2008. The role of
social capital in market opportunity identification for emerg-
ing technologies. Working paper CDM-EPFL & Sol C. Snider
Entrepreneurial Research Center, Wharton School, University
of Pennsylvania, Philadelphia.
Halford, G., W. Wilson, J. Guo, et al. 1994. Connectionist impli-
cations for processing capacity limitations in analogies. R. W.
Gayler, J. Wiles, J. E. M. Stewart, eds. Advances in Connection-
ist and Neural Computation Theory, Vol. 2, Ablex, Norwood, NJ,
363–415.
Hamilton, B. H., J. A. Nickerson. 2003. Correcting for endogeneity
in strategic management research. Strategic Organ. 151–78.
Hausman, J., B. H. Hall, Z. Griliches. 1984. Econometric models for
count data with an application to the patents-R&D relation-
ship. Econometrica 52 909–938.
Heckman, J. J. 1979. Sample selection bias as a specification error.
Econometrica 47 153–161.
Helfat, C. E., M. B. Lieberman. 2002. The birth of capabilities: Mar-
ket entry and the importance of pre-history. Indust. Corporate
Change 11 725–760.
Huber, G. P., D. J. Power. 1985. Retrospective reports of strategic-
level managers: Guidelines for increasing their accuracy. Strate-
gic Management J. 6171–180.
Janis, I. L., L. Mann. 1977. Decision Making: A Psychological Analysis
of Conflict, Choice, and Commitment. Free Press, New York.
Jolly, V. K 1997. Commercializing New Technologies: Getting from Mind
to Market. Harvard Business School Press, Boston.
Katila, R., G. Ahuja. 2002. Something old, something new: A lon-
gitudinal study of search behavior and new product introduc-
tion. Acad. Management J. 45 1183–1194.
Knight, F. H. 1921. Risk, Uncertainty and Profit. Houghton Mifflin,
New York.
Knudsen, T., D. A. Levinthal. 2007. Two faces of search: Alternative
generation and alternative evaluation. Organ. Sci. 18 39–54.
March, J. G. 1991. Exploration and exploitation in organizational
learning. Organ. Sci. 271–87.
March, J. G., H. Simon. 1958. Organizations. Wiley, New York.
McGrath, R. G., I. C. MacMillan. 2000. The Entrepreneurial Mindset.
Harvard Business School Press, Boston.
McKnight, T. 2004. Will It Fly?How to Know If Your Business Idea
Has WingsBefore You Take the Leap. Prentice Hall, Englewood
Cliffs, NJ.
Miller, D. J. 2006. Technological diversity, related diversification,
and firm performance. Strat. Management J. 27 601–619.
Miller, D. J., M. J. Fern, L. B. Cardinal. 2007. The use of knowledge
for technological innovation within the diversified firm. Acad.
Management J. 50 308–326.
Mitchell, W. 1992. Are more good things better, or will technical
and market capabilities conflict when a firm expands? Indust.
Corporate Change 1327–346.
Mullins, J. W. 2003. The New Business Road Test: What Entrepreneurs
and Executives Should Do Before Writing a Business Plan.
FT/Prentice Hall, London.
Nerkar, A., P. W. Roberts. 2004. Technological and product-market
experience and the success of new product introductions in the
pharmaceutical industry. Strategic Management J. 25 779–799.
Osborn, A. F. 1957. Applied ImaginationPrinciples and Procedures of
Creative Thinking. Scribner, New York.
Pelled, L. H., K. M. Eisenhardt, K. R. Xin. 1999. Exploring the black
box: An analysis of work group diversity, conflict, and perfor-
mance. Admin. Sci. Quart. 44 1–28.
Penrose, E. T. 1959. The Theory of the Growth of the Firm. Oxford
University Press, Oxford, UK.
Peteraf, M. A., M. E. Bergen. 2003. Scanning dynamic competitive
landscapes: A market-based and resource-based framework.
Strategic Management J. 24 1027–1041.
Peters, T. J., R. H. Waterman. 1982. In Search of Excellence. Harper
and Row, New York.
Prahalad, C. K., G. Hamel. 1990. The core competence of the cor-
poration. Harvard Bus. Rev. 68(3) 79–91.
Radner, R. 1996. Bounded rationality, indeterminacy, and the theory
of the firm. Econom. J. 106 1360–1373.
Rosenkopf, L., A. Nerkar. 2001. Beyond local search: Boundary-
spanning, exploration, and impact in the optical disk industry.
Strategic Management J. 22 287–306.
Roy, P. 2005. Leveraging existing technology: The role of alliances in
the cross-application across technology domains in high-tech
entrepreneurial firms. Working paper, Wharton School, Uni-
versity of Pennsylvania, Philadelphia.
Shane, S. 2000. Prior knowledge and the discovery of
entrepreneurial opportunities. Organ. Sci. 11 448–469.
Shane, S. 2004. Academic EntrepreneurshipUniversity Spinoffs and
Wealth Creation. Elgar, Cheltenham, UK.
Shane, S., S. Venkataraman. 2000. The promise of entrepreneurship
as a field of research. Acad. Management Rev. 24 217–226.
Simon, H. A. 1982. Models of Bounded Rationality. MIT Press,
Cambridge, MA.
Stuart, T. E., J. M. Podolny. 1996. Local search and the evolution
of technological capabilities. Strat. Management J. 17(Summer
Special Issue) 21–38.
Taylor, A., H. R. Greve. 2006. Superman or the fantastic four?
Knowledge combination and experience in innovative teams.
Acad. Management J. 49 723–740.
Westhead, P., M. Wright. 1998. Novice, portfolio, and serial
founders: Are they different? J. Bus. Venturing 13 173–204.
Wiersema, M. F., K. A. Bantel. 1992. Top management team demog-
raphy and corporate strategic change. Acad. Management J. 35
91–121.
Wooldridge, J. M. 2002. Econometric Analysis of Cross Section and
Panel Data. MIT Press, Cambridge, MA.