(175(35(1(85,$//($51,1*352&(66(6,1

$&$'(0,&63,12))6
Petra Dickel, Universitv of Kiel, Germanv
Anke Rasmus, Universitv of Kiel, Germanv
Achim Walter, Universitv of Kiel, Germanv
$%675$&7
Academic spin-oIIs are an important means Ior technology transIer Irom public research and strongly
contribute to regional development. However, these companies have to overcome major challenges at start-up.
Academic Iounders oIten lack business know-how and a market-oriented way oI thinking. At the same time
spin-oII technology oIten still is at an early development stage and involves considerable uncertainties.
Accordingly, many spin-oIIs Iollow suboptimal market chances and Iail. SuccessIul entrepreneurs, on the
contrary, learn about markets right Irom the start and incorporate this knowledge in their product development
activities. So Iar, little is known about the nature oI these market-based learning processes in new ventures.
Drawing on a database oI 71 academic spin-oIIs, this study investigates how new ventures learn about
markets at an early stage and how this inIluences market perIormance oI spin-oIIs.
,1752'8&7,21
The role oI universities and public research institutions in transIerring innovative technologies is widely
acknowledged. Technology transIer increasingly becomes important with the change Irom industry to
knowledge based economies with respect to technological progress and economic growth. In the context oI
academic institutions, diIIerent channels oI technology transIer exist, e.g. scientiIic publications, cooperation
and contract research, consulting or licensing. Above all, spin-oIIs are regarded to particularly contribute to
regional economic development (DorIman, 1983; Chrisman et al., 1995; Lockett et al., 2003) and hence have
been subject oI increased interest in the past years. Academic spin-oIIs are companies that are Iounded by
employees oI public research institutions in order to commercialise technology developed within the
institution (SteIIensen et al., 1999).
However, in order to realise positive economic eIIects spin-oIIs must successIully innovate, i.e. transIer
their technology into marketable products. This poses a major challenge Ior a lot oI spin-oIIs (Vohora et al.,
2004). Spin-oII Iounders oIten lack business knowledge and a market-oriented way oI thinking (KloIsten et
al., 1988). At the same time, spin-oII technology oIten is still at an early development stage, i.e. Iar Irom a
possible commercialization and requires Iurther development (Jensen & Thursby, 2001). Thus, it is oIten
ambiguous which product solutions are most appropriate. Furthermore, a single technology can serve diIIerent
markets diIIerently well. This means that academic spin-oIIs can enter many diIIerent markets with their
technology, but only Iew oI those market chances possess high growth potential (Shane, 2000; Ardichvili et
al., 2003). Hence, the decision on a speciIic target market is crucial to spin-oII success. In this context,
traditional market research methods are ill-suited to gain realistic market data and insights due to several
reasons. It is not only the identiIication oI relevant users that already poses a problem at this early stage, also
user needs are mostly latent and thereIore diIIicult to articulate. Besides, concept and prototype evaluation is
hard Ior users in the light oI missing reIerence products and experience (Von Hippel, 1986; Lettl, 2004).
Accordingly, many spin-oIIs Iollow suboptimal market chances without having a clear vision oI relevant
target markets and user requirements and Iail accordingly.
Lynn and Akgn (1998) argue that a learning-based strategy is appropriate in such an environment
characterised by high uncertainties. However, only little is known how learning processes take place in new
ventures. Particularly, there is Iew empirical evidence on how Iounders make sense about the oIten ambiguous
markets and customer needs at an early stage. This study`s objective is to close this gap by investigating
market-based learning processes in academic spin-oIIs as well as iI there are diIIerences in learning between
successIul and unsuccessIul spin-oIIs.
In this article we will Iirst present major theoretical Iindings Irom which main hypotheses are derived.
Then, methodology and results oI an empirical study oI 71 academic spin-oIIs will be presented. We conclude
with managerial implications and Iuture research questions.
Page 1 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
THEORY & MODEL
Market-Based Organizational Learning & Entrepreneurship
Organizational learning is 'a process oI improving actions through better knowledge and
understanding (Fiol & Lyles, 1985, p. 803). Learning Iosters innovation, reduces the risk oI knowledge
becoming obsolete and hence contributes to sustaining a company`s dynamic capabilities (Teece et al., 1997).
The concept oI learning is particularly discussed among strategic management and organization research
schools, e.g. in the context oI organizations (Fiol & Lyles, 1985; Huber, 1991; March, 1991; Nonaka, 1994)
and alliances (Lane & Lubatkin, 1998; Larsson et al., 1998; Inkpen, 2000), learning barriers (Kim, 1993;
Szulanski, 1996) or the risk oI outlearning and spill-over eIIects (Hamel et al., 1989; Hamel, 1991). However,
there is neither a common notion nor term oI learning yet. In this paper, learning is conceptualized as a multi-
stage process consisting oI knowledge acquisition and knowledge integration activities (Day, 1994; Zahra et
al., 2000). Knowledge is the result oI learning processes and can be distinguished in explicit and tacit
knowledge. Explicit knowledge can be codiIied in the Iorm oI language, numbers, symbols and Iigures and is
thereIore easy to transIer. Tacit knowledge dependents on context and person, cannot be articulated and is
hence diIIicult to transIer (Polanyi, 1958; Nonaka, 1994).
With respect to market-based learning, there is a wide consensus on the positive relationship between
market and customer understanding and perIormance (e.g. Dougherty, 1990; Kohli & Jaworski, 1990; Narver
& Slater, 1990; Sinkula, 1994; Li & Calantone, 1998). However, existing studies concentrate on established
companies that learn about changes in their business environment and adapt to those changes. The business
environment oI academic spin-oIIs, on the contrary, is not yet deIined and diIIerent in many aspects. First,
spin-oIIs can potentially enter many diIIerent markets with one single technology (Shane, 2000), which places
the emphasis on the deIinition and choice oI relevant target markets rather than the introduction oI
incremental innovations in existing markets. Second, new ventures cannot Iall back on existing knowledge
bases and routines, but rather have to build them up in order to survive and grow. Third, coordination tasks oI
diIIerent organizational Iunctions and departments (e.g. Dougherty, 1990; Narver & Slater, 1990) are less
problematic in new ventures` learning attempts but rather the question how and Irom where relevant market
knowledge can be acquired and how this knowledge can be integrated in new product development. Spin-oII`s
learning requirements thereIore signiIicantly diIIer Irom those oI established companies both in complexity as
well as their direct and long-term consequences on business development.
Although the importance oI learning is widely acknowledged in entrepreneurship research, learning in the
context oI start-ups is still ill-understood (Harrison & Leitch 2005). Most studies Iocus on the individual level
oI the Iounder, i.e. her behavior and cognitive abilities (e.g. Minniti & Bygrave, 2001; Corbett, 2005).
Empirical research on entrepreneurial learning at an organizational level is rare and appears relatively
heterogeneous (Zahra et al., 2006). Studies show that entrepreneurs usually experiment with diIIerent business
models and successively learn about their environment (Nicholls-Nixon et al., 2000; Ravasi & Turati,
2005).Zahra et al. (2000) demonstrate the positive perIormance eIIect oI technological learning gained
through internationalization oI new ventures. Yli-Renko et al. (2001) analyzed knowledge acquisition Irom
key customers; their results illustrate the relevance oI strong personal contacts in entrepreneurial learning
processes. However, organizational learning is oIten conceptualized very generally, e.g. as a Iunction oI
knowledge access structures (Almeida et al., 2003) or the degree oI organizational change (e.g. Nicholls-
Nixon et al., 2001), which hardly contribute to the understanding oI the underlying learning processes. The
case studies oI Xie and White (2004) and Ravasi and Turati (2005) deliver greater in-depth insights by
comparing the learning processes in one start-up each. However, the studies` transIerability is questionable
due to their very speciIic context oI one particular company. Peters and Brush (1996) give diIIerentiated
insights oI new ventures` market inIormation activities but their descriptive study design does not enable to
deduce causal eIIects. Moreover, many studies oI entrepreneurial learning Iocus on technological learning, i.e.
the acquisition and integration oI new technological skills (Zahra et al., 2000; Saemundsson, 2005). Market-
based learning, however, is rarely investigated, which is rather surprising keeping in mind the importance oI
market inIormation especially Ior young companies` survival and growth (Bullinger, 1990; Hemer, 2006).
Vohora et al. (2004) conclude Irom their case studies that spin-oIIs must possess at least some market
understanding to survive. Bond and Houston (2003, p. 132) argue a same and point out that 'market-
technology barriers that we identiIied highlight the importance oI the Iirm`s attention to collecting
inIormation regarding technological possibilities, customer needs, competitors` technology strategy and
market Ieasibility as inputs to making strategic technology decisions.'
This paper`s objective is to build on and extend existing research by analyzing market-based learning
Page 2 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
processes in academic spin-oIIs. We particularly investigate how academic spin-oIIs learn about markets at
an early stage and how this inIluences their market perIormance. We diIIerentiate between personal and
codiIied learning as distinct knowledge acquisition strategies and analyze how spin-oIIs integrate market
knowledge in their business activities. We also investigate whether the impact oI these market-based learning
processes change in dynamic environments.
Our study contributes to entrepreneurial learning research by closely analyzing learning processes in
academic spin-oIIs. As well it compares the impact oI diIIerent learning strategies with respect to gaining
market knowledge and shows how dynamic environments eIIect a spin-oII`s learning eIIorts. It Iurther
contributes to organizational learning research by extending market-based learning to the context oI new
ventures.
Figure 1 shows the study`s basic research model. According to this, spin-oII`s perIormance is inIluenced by
knowledge acquisition in Iorm oI personal (PL) and codiIied learning (CL) as well as knowledge integration
(KI) activities. A moderating eIIect oI technological dynamism (DYN) is assumed. Market perIormance (MP)
is captured by the spin-oII`s Iirst product success indicating how successIul the company was in transIerring
technology into a marketable product. Constructs and relationships will be discussed in the Iollowing
section.
3HUVRQDODQG&RGLILHG/HDUQLQJ
According to the knowledge based view, knowledge is a company`s most important resource and
perIormance is strongly determined by the ability to apply this knowledge eIIectively (Penrose, 1959; Grant,
1996a). In the case oI academic spin-oIIs, it is necessary to connect their technological know-how with
speciIic market demands (Bond & Houston, 2003). Thus, spin-oIIs must be able to translate their technology
into marketable products by gaining and integrating market knowledge in their product development
activities.
Knowledge can be acquired personally or through codiIied sources, i.e. independent oI other persons (DaIt
& Weick, 1984). Personal learning (PL) reIers to the acquisition oI inIormation and skills that are closely
connected to other persons and hence necessitates close personal cooperation (Peters & Brush, 1996; Hansen
et al., 1999). Both potential users (Von Hippel, 1986), suppliers and competitors (Baum et al., 2000; Vohora
et al., 2004) as well as independent industry experts (Vohora et al., 2004) can be appropriate sources Ior
market inIormation. In the context oI academic spin-oIIs PL plays a pivotal role since relevant market
knowledge usually is not publicly available and concentrated on only a Iew people (Zucker et al., 2002).
Moreover, user needs are mostly latent and thereIore oIten cannot be articulated (Von Hippel, 1986). This
highly tacit knowledge can only be transIerred via personal interaction with potential users and market
experts, in order to make use oI it in spin-oIIs' product development activities (Nonaka, 1994; Lane &
Lubatkin, 1998). Thus, a personal based knowledge acquisition is vital Ior academic spin-oIIs in order to
access the mainly tacit knowledge about markets and user needs. We thereIore assume a positive relationship
between PL and market perIormance.
H
1
. The extent of personal learning positivelv influences a spin-offs market performance. The higher the
extent of knowledge acquisition through personal sources during product development the higher a spin-
offs market performance.
CodiIied learning (CL) reIers to inIormation acquisition independent oI other people through codiIied
sources (Peters & Brush, 1996; Hansen et al., 1999). This explicit knowledge is tradable, e.g. in Iorm oI
traditional market research data and research publications, or is publicly accessible, e.g. in Iorm oI market
data published in internet and print media. Due to its higher transIerability CL involves less time and eIIort in
accessing knowledge than PL. However, the value oI codiIied inIormation acquired is lower since this
inIormation is potentially available to any other party, such as competitors, too. As well it only allows gaining
superIicial general insights (Lane & Lubatkin, 1998). However, we believe that CL complements PL through
the acquisition oI basic know-how, such as legal requirements and general market trends. Thus, we also
expect positive eIIect oI codiIied learning on market perIormance.
H
2
. The extent of codified learning positivelv influences a spin-offs market performance. The higher the
extent of knowledge acquisition through codified sources during product development the higher a spin-
offs market performance.
Page 3 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
.QRZOHGJH,QWHJUDWLRQ
In order to become eIIective, knowledge gained needs to be integrated within the organization (Cohen &
Levinthal, 1990; Grant, 1996b). Knowledge acquisition and knowledge integration are thereIore
complementary to each other and deIine an organization`s learning process.
Knowledge integration (KI) is a three-step process that describes how the knowledge gained is captured,
interpreted and used in the company (Zahra et al., 2000). Capture reIers to the documentation and exchange
oI inIormation within the spin-oII in order to make it available to relevant organizational members.
Interpretation is necessary to reIlect and evaluate new inIormation. This takes place by jointly discussing
analyzing successIul and less successIul project activities as well as a rather trial-and-error based inIormation
processing by which a company successively gains clarity on what has been learnt (Von Hippel & Tyre, 1995;
Lynn et al., 1996). Finally new knowledge must be put to use, i.e. applied in new product development.
ThereIore it is mandatory to draw the consequences oI the new market inIormation gained and to integrate this
knowledge into the spin-oII`s business activities. In turn, we expect that KI positively inIluences a spin-oII`s
market perIormance.
H
3
. Knowledge integration describes the capture, interpretation and use of market information acquired.
The extent of knowledge integration positivelv influences a spin-offs market performance. The higher the
extent of knowledge integration during product development the higher a spin-offs market performance.
7HFKQRORJLFDO'\QDPLVP
According to Glazer and Weiss (1993) a market with environmental turbulence is characterized by (1) a
high extent oI inter-period change in the levels or values oI key environmental Iactors and (2) substantial
uncertainty and unpredictability as to the Iuture values oI these variables.
The impact oI dynamic environments is mainly discussed in marketing and strategy research. Jaworski and
Kohli (1993) empirically analyzed the relationship between market orientation and business perIormance and
the impact oI environmental variables that comprise varying degrees oI market turbulence, competitive
intensity and technological turbulence. Their results did not show any signiIicant eIIects oI environmental
Iactors on the market orientation perIormance relationship. Similarly, Calantone et al. (2003) Iound little
support Ior an interaction eIIect between turbulent environments and market orientation on perIormance. The
empirical study oI Han et al. (1998) indicates some diIIerent results. Though they Iound no inIluence oI
market turbulence, they show that technological turbulence moderates the relationship between market
orientation and innovation development. Thus, the impact oI dynamic environments with respect to market
understanding remains ambiguous. However, as these studies Iocus on established companies, the situation
might be quite diIIerent Ior technology-based new ventures, in which the eIIectiveness oI market-based
learning is probably more prone to the degree oI technological dynamism.
Academic spin-oIIs, as most technology based companies, are usually situated in highly dynamic and
volatile environments. This implies that time plays a major role in this context that is characterized by an
acceleration oI market and technological changes and an explosion oI available inIormation (Slater & Narver,
1995). In these environments, spin-oIIs have to cope with an abundance oI new emerging market and
technological data within short time while they develop and commercialize their technology. Bourgeois and
Eisenhardt (1988) describe highly dynamic markets in which changes are so rapid and discontinuous that
inIormation collection at the beginning oI new product development cycle can become obsolete by the time
products are launched. Thus, market inIormation becomes time sensitive in these environments (Glazer &
Weiss, 1993). An immediate and direct access to market knowledge creates a competitive advantage due to
the reduction oI development time oI innovations (Calantone et al., 2003) and the ability to earlier launch and
market new products than competitors. This is crucial in turbulent environments, in which technological liIe
cycles become shorter and technological breakthroughs occur at a more rapid pace. Hence, the capability to
respond rapidly to evolving technical and market inIormation is a critical Iactor to success (Iansiti, 1995). As
competitive advantage oI spin-oIIs strongly depends on their technology and the ability to eIIectively
commercialize it, this study particularly Iocuses on the impact oI technological dynamism (DYN). This is
deIined as a turbulent environment, in which the time intervals between major technological innovations
continuously decrease.
Page 4 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
Personal learning is particularly important in dynamic environments, as the highly tacit market knowledge
can only be transIerred in short time via personal interaction with external partners. Close cooperation with
users, suppliers, and market experts hence is necessary to stay abreast oI competitors. TransIerring
inIormation via personal interaction allows accessing up-to-date market inIormation within short time.
Moreover, the direct contact with external partners does not only enable to immediately request and transIer
knowledge, the interacting parties can also instantly Ieedback on the inIormation exchanged and thereby
reduce misunderstandings. Accordingly we assume that an environment characterized by high technological
dynamism reinIorces the relationship between personal learning and spin-oII`s market perIormance.
H
4a
. The extent of technological dvnamism positivelv moderates the effect of personal learning on market
performance of academic spin-offs.
As codiIied learning builds on knowledge acquisition Irom existing and accessible data and publications,
this learning strategy involves the risk that the knowledge sources are less up-to-date than personal sources.
Indeed, codiIied knowledge is characterized by some time lag reIlected in the time between the discovery,
processing, storage and publication oI inIormation and data. Especially market research data, iI available, is
time-delayed as it rather comprises past customer needs and competitor activities than Iuture customer
attitudes and requirements and competitor strategies respectively. Contrary to personal learning, we assume
that a high extent oI technological dynamism weakens the eIIect oI codiIied learning on spin-oII`s market
perIormance.
H
4b
. The extent of technological dvnamism negativelv moderates the effect of codified learning on market
performance of academic spin-offs.
Concerning knowledge integration, spin-oIIs must respond rapidly to new market inIormation. This is
especially the case in highly dynamic environments, in which the ability to react Iaster than competitors is
crucial. II knowledge is not captured, interpreted and used in short time, superior knowledge will become
obsolete and competitive advantage will be lost soon. Accordingly we expect that the relationship between
knowledge integration and market perIormance is reinIorced by technological dynamism.
H
4c
. The extent of technological dvnamism positivelv moderates the effect of knowledge integration on
market performance of academic spin-offs.
0(7+2'2/2*<
To examine the relationship between market learning processes and market perIormance, a study oI 152
spin-oIIs Irom higher education institutions in Germany was conducted. To be included in the sample Iour
criteria had to be met: (1) the spin-oII has been Iounded by employees oI a public research institution, (2) the
spin-oII`s core technology originates Irom this research institution, (3) the company is independent and (4) its
maximum age is 10 years.
Data was collected in Iorm oI standardized questionnaires in personal interviews with spin-oII Iounders. A
multiple key inIormant approach was used as suggested by Kumar et al. (1993) and Ernst (2002) to reduce a
possible inIormant bias. The spin-oII`s Iounder was ask to state on the spin-oII`s characteristics at start-up as
well as perIormance while a second key inIormant who was involved in new product development and launch
(e.g. project manager) was ask in a shorter questionnaire to indicate the knowledge acquisition and integration
activities during that time. The Iinal sample consists oI 116 usable questionnaires oI which 71 were Iully
completed by both key inIormants. These spin-oIIs have an average age oI 5 years, size oI 11,5 employees
and turnover oI 887 Tt in 2005 which is comparable to other studies on academic spin-oIIs (Jones-Evans et
al., 1998; SteIIensen et al., 1999). 37 oI the companies are based on inIormation technology, 21 on
biotechnology, 23 on electronics, photonic and nanotechnology and 19 on other technological Iields.
Scales were either adapted Irom literature or developed speciIically Ior this study. Guidelines set by
Diamantopolous and WinklhoIer (2001) were Iollowed in developing the Iormative scales. Initial item pool
generation is based on in-depth interviews with spin-oII Iounders. Scales were successively pretested and
revised aIter each round. At the end oI the third Iinal round, respondents indicated that the remaining items
were clear, meaningIul and relevant. All items were measured on a seven-point Likert scale.
Personal learning was captured by Iive items on market knowledge acquisition activities by interacting
Page 5 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
with diIIerent personal knowledge sources (e.g. customers, competitors, market experts) during new
product development. Similarly, Iive items reIerred to codified learning in terms oI knowledge acquisition
Irom codiIied sources (e.g. databases, magazines, internet). Knowledge integration was measured with eleven
items on the capture, interpretation and use oI market inIormation gained during new product development.
Technological dvnamism was captured by one item that described the extent oI decreasing time intervals
between major technological innovations in the industry.
To evaluate performance, many empirical studies use Iinancial (objective) measures, as e.g. sales, proIit,
cash-Ilow etc. (Geringer & Herbert 1991). Since this data is very oIten diIIicult to collect due to
conIidentiality reasons or non-availability oI data which is particularly the case Ior new ventures
researchers increasingly use perIormance indicators that are based on subjective evaluations oI key
inIormants. A lot oI studies show that these subjective measures highly correlate with objective criteria and
are thereIore an appropriate option Ior perIormance measurement (e.g. Covin & Slevin 1989; Geringer &
Herbert, 1991; Kale et al., 2002). In our study, a subjective perIormance measure was used as dependent
variable. Market performance was measured with seven items that indicate how the spin-oII`s Iirst product
perIormed one year aIter launch in terms oI the achievement oI Iounders` objectives. In order to reduce
inIormant bias, items on learning processes and market perIormance were evaluated by two diIIerent key
inIormants in separate questionnaires as described above.
We controlled Ior prior knowledge in terms oI the Iounders` industry experience in years as well as
existing relationships with customers at start-up. Moreover, spin-oII`s main technological Iield, Iounders'
team size and technological uncertainty at start up were used as controls. Finally we controlled Ior the time
lag between the spin-oII`s start-up and their Iirst product launch.
(03,5,&$/5(68/76
To test the hypotheses a moderated regression analysis was perIormed as suggested by Aiken and West
(1991). According to this, all independent variables are simultaneously considered in a step-wise process in
which Iirst the controls, then the main eIIect variables and Iinally the interaction eIIects are included into the
regression equation. All independent variables were centred prior the computation oI interaction terms.
All measures were analyzed Ior reliability and validity. The correlations (Pearson) oI the main variables
ranged Irom 0.02 to 0.46. Descriptive statistics and correlations oI main variables are shown in table 1.
Moderated regression analysis as shown in table 2 conIirmed the proposed direct positive eIIects oI
personal learning and knowledge integration on perIormance (H
1
and H
3
supported), while there the direct
eIIect oI codiIied learning is not signiIicant (H
2
not supported). However, our results indicate a positive
interaction eIIect oI personal learning and technological dynamism and a negative interaction oI codiIied
learning and technological dynamism on perIormance (H
4a
and H
4b
supported). The eIIect oI knowledge
integration does not change in technologically dynamic environments (H
4c
not supported).
Further tests indicate that multicollinearity is unlikely to pose a problem with variance inIlations Iactors
(VIF) below 1.8 and a maximum condition index (CI) oI 3.9. Multicollinearity possibly exists iI correlations
are above 0.70, VIFs above 10, and CI is higher than 30. Moreover, a multi-trait multi-method (MTMM)
analysis as proposed by Campbell and Fiske (1959) was conducted to control Ior validity oI key inIormants`
statements. MTMM analysis showed that both convergent validity and discriminant validity oI key
inIormants` statements can be conIirmed.
To Iurther analyse results the signiIicant interaction eIIects were graphically plotted as proposed by Aiken
and West (1991). Figure 2 shows the interaction oI personal learning and technological dynamism on
perIormance. Personal learning obviously is oI minor importance when a spin-oII`s environment is less
dynamic but it becomes a crucial Iactor with increasing technological dynamism. On the contrary, codiIied
learning becomes increasingly counterproductive the more a spin-oII`s environment is characterised by high
technological dynamism as shown in Iigure 3.
',6&866,21$1',03/,&$7,216
Summing up, our study shows a positive impact oI personal learning and knowledge integration on spin-
Page 6 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
oII`s market perIormance. Results Iurther indicate a positive interaction eIIect oI personal learning and
technological dynamics on perIormance, while codiIied learning has a negative eIIect in technologically
dynamic environments. We conclude that personal learning and knowledge integration are important means
Ior developing marketable products in academic spin-oIIs, while codiIied learning seems to be a less eIIective
or even a counterproductive way to acquire market inIormation at an early stage.
Our results show the importance oI investigating the learning concept in a more diIIerentiated way. A
personal based learning strategy seems to be particularly appropriate in the context oI spin-oIIs; also
knowledge integration is not to be neglected when researching learning processes. Results also indicate that
studies should go beyond the prevailing user perspective. Other market participants such as suppliers,
competitors etc. can also be sources Ior valuable market inIormation Ior new ventures. Finally, learning seems
to be diIIerently important in diIIerent environments, hence environmental Iactors such as e.g. technological
dynamism are also be taken into account when deIining a spin-oII`s appropriate learning strategy.
From a managerial perspective the results underscore the necessity that academic spin-oIIs must learn
about markets already at an early stage in order to realise marketable products and to succeed in the market
place. However, there are diIIerent ways oI learning and their impact diIIers signiIicantly. Concerning
knowledge acquisition a person-based way oI learning is more eIIective than studying inIormation Irom
codiIied sources. Thus, the early integration oI potential customers and suppliers in product development is
mandatory as well as the close interaction with market experts Irom the start. Comprehensive exchange with
diIIerent market partners is crucial to gain a sense Ior markets and close collaboration is necessary to get
valuable market inIormation that otherwise could not be acquired. Market research or any other public
available sources are less eIIective devices to learn about markets at an early stage. This is particularly the
case in technologically dynamic environments in which current publicized data is not up-to-date any more
within a short period oI time. Moreover, comprehensive knowledge integration is a precondition to realize
perIormance eIIects. For this purpose academic Iounders should establish processes that Iacilitate knowledge
Ilows within the organization and set up structures that Ioster open communication and constructive dialogue
between organizational members in order to Iully tap the potential oI what has been learnt.
Although this study provides some interesting results, some limitations are to be mentioned. Our study
used a subjective perIormance measurement, which might give rise to biased responses. As our data builds on
two key inIormants we managed to minimize the risk oI an inIormant bias, at least to some extent, which is
also conIirmed by the MTMM analysis that indicates validity oI results. Further, our data derives Irom a
cross-sectional study. As learning is inherently dynamic in nature, e.g. relevant market knowledge may
change over time according to current needs and requirements, Iuture research should investigate the
development oI learning processes in spin-oIIs and their perIormance impact in a longitudal study. Finally, the
role oI Iurther context Iactors on the eIIectiveness oI entrepreneurial learning would be a promising Iield oI
research. Future studies should thereIore investigate the inIluence oI spin-oIIs` diIIerent start-up
conIigurations in terms oI their resources and entrepreneurial culture to complete our understanding oI
learning processes in academic spin-oIIs.
&217$&7 Petra Dickel; dickelbwl.uni-kiel.de; (T): 49-431-8804698; (F): 49-431-8803213; Christian-
Albrechts-University Kiel, Institute Ior Innovation Research, Westring 425, 24118 Kiel, Germany.
5()(5(1&(6
Aiken, L.S., West, S.G. (1992): Multiple regression. Testing and interpreting interactions. Sage. Newbury
Park, CaliIornia.
Almeida, P., Dokko, G., RosenkopI, L. (2003): Startup size and the mechanisms oI external learning:
Increasing opportunity and decreasing ability? Research Policv 32 (2), p. 301-315.
Ardichvili, A., Cardozo, R., Ray, S. (2003): A theory oI entrepreneurial opportunity identiIication and
development. Journal of Business Jenturing, 18 (18), p. 105-123.
Baum, J.A.C., Calabrese T., Silverman B. (2000): Don't go it alone: Alliance network composition and
startups' perIormance in Canadian biotechnology. Strategic Management Journal, 21(3), p. 267-294.
Bond, E.U., Houston, M.B., (2003): Barriers to matching new technologies and market opportunities in
established Iirms. Journal of Product Innovation Management, 20 (2), p. 120-135.
Bourgeois, L.J., Eisenhardt, K.M. (1988): Strategic decision processes in high velocity environments: Iour
cases in the microcomputer industry. Management Science, 34 (7), p. 816-835.
Bullinger, H.-J. (1990): F&E - heute. Industrielle Forschung und Entwicklung in der Bundesrepublik
Deutschland. IAO-Studie, Mnchen: gImt.
Page 7 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
Calantone, R. Garcia, R., Droege, C. (2003): The eIIects oI environmental turbulence on new product
development strategy planning. Journal of Product Innovation Management, 20 (2), p. 90-103.
Campbell, D.T., Fiske, D.W. (1959): Convergent and discriminant validation by the multitrait-multimethod
matrix. Psvchological Bulletin, 56 (2), p. 81-105.
Chrisman, J.J., Hynes, T., Fraser, S. (1995): Faculty entrepreneurship and economic development: the case oI
the university oI Calgary. Journal of Business Jenturing, 10 (4), p. 267-281.
Cohen, W.M., Levinthal, D.A. (1990): Absorptive capacity: A new perspective on learning and innovation.
Administrative Science Quarterlv, 35 (1), p. 128-152.
Corbett, A.C. (2005): Experiential learning within the process oI opportunity identiIication and exploitation.
Entrepreneurship Theorv & Practice, 29 (4), p. 473-492.
Covin, J.G., Slevin, D.P. (1989): Strategic management oI small Iirms in hostile and benign environments.
Strategic Management Journal, 10 (1), p. 75-87.
DaIt, R., Weick, K. (1984): Toward a model oI organizations as interpretation systems. Academv of
Management Review, 9 (2), p. 284-296.
Day, G.S. (1994): Continuous learning about markets. California Management Review, 36 (4), p. 9-31.
Diamantopoulos, A., WinklhoIer, H.M. (2001): Index construction with Iormative indicators: An alternative
to scale development. Journal of Marketing Research, 38 (2), p. 269-277.
DorIman, N.S. (1983): Route 128: The development oI a regional high technology economy. Research Policv,
12 (6), p. 299-316.
Dougherty, D. (1990): Understanding new markets Ior new products. Strategic Management Journal, 11 (4),
p. 59-78.
Ernst, H. (2002): Success Iactors oI new product development: A review oI the empirical literature.
International Journal of Management Review, 4 (1), p. 1-40.
Fiol, C.M., Lyles, M.A. (1985): Organizational learning. Academv of Management Review, 10 (4), p. 803-813.
Geringer, J.M., Herbert, L. (1991): Measuring perIormance oI international joint ventures. Journal of
International Business Studies, 22 (2), p. 249-264.
Glazer, R., Weiss, A.M. (1993): Marketing in turbulent environments: Decision processes and time-sensitivity
oI inIormation. Journal of Marketing Research, 15 (4), p. 509-521.
Grant, R.M. (1996a): Towards a knowledge-based view oI the Iirm. Strategic Management Journal, 17
Special Issue, p. 109-122.
Grant, R.M. (1996b): Prospering in dynamically-competitive environments: Organizational capability as
knowledge integration. Organi:ation Science, 7 (4), p. 375-387.
Hamel, G. (1991): Competition Ior competence and inter-partner learning within international strategic
alliances. Strategic Management Journal, 12 (4), p. 83-103.
Hamel, G., Doz, Y.L., Prahalad C.K. (1989): Collaborate with your competitors and win. Harvard Business
Review, 67 (1), p. 133-139.
Han, J.K., Kim, N., Srivastava, R.K. (1998): Market orientation and organizational perIormance: Is innovation
a missing link? Journal of Marketing, 62 (4), p. 30-45.
Hansen, M.T., Nohria, N., Tierney, T. (1999): What's your strategy Ior managing knowledge? Harvard
Business Review, 77 (2), p. 106-116.
Harrison, R.T., Leitch, C.M. (2005): Entrepreneurial learning: Researching the interIace between learning and
the entrepreneurial context. Entrepreneurship Theorv & Practice, 29 (4), p. 351-371.
Hemer, J., Berteit, H., Walter, G., Gthner, M. (2006): Success Factors for Academic Spin-offs. Stuttgart:
FraunhoIer IRB Verlag.
Huber, G.P. (1991): Organizational learning: The contributing processes and the literatures. Organi:ation
Science, 2 (1), p. 88-115.
Iansiti, M. (1995): Shooting the rapids: Managing product development in turbulent environments. California
Management Review, 38 (1), p. 37-58.
Inkpen, A.C. (2000): Learning through joint ventures: A Iramework oI knowledge acquisition. Journal of
Management Studies, 37 (7), p. 1022-1043.
Jaworski, B.J., Kohli, A.K. (1993): Market orientation: Antecedents and consequences. Journal of Marketing,
57 (3), p. 53-70.
Jensen, R.J., Thursby, M. (2001): ProoIs and prototypes Ior sale: The licensing oI university inventions. The
American Economic Review, 91 (1), p. 240-259.
Jones-Evans, D. (1997): Technical entrepreneurship, experience and the management oI small technology-
based Iirms. in: Jones-Evans, D., KloIsten, M. (eds.): Technologv, innovation and enterprise. The
European experience, p. 11-60.
Kale, P., Dyer, J.H., Singh, H. (2002): Alliance capability, stock market response and long-term alliance
success: The role oI the alliance Iunction. Strategic Management Journal, 23 (8), p. 747-767.
Kim, D.H. (1993): The link between individual and organizational learning. Sloan Management Review, 35
Page 8 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
(1), p. 37-50.
KloIsten, M., Lindell P., OloIIson, C., Wahlbin, C. (1988): Internal and external resources in technology-
based spin-oIIs. Frontiers of Entrepreneurship Research, p. 430-433.
Kohli, A.K., Jaworski, B.J. (1990): Market orientation: The construct, research propositions, and managerial
implications. Journal of Marketing, 54 (2), p. 1-18.
Kumar, N., Stern, L., Anderson, J. (1993): Conducting interorganisational research using key inIormants.
Academv of Management Journal, 36 (6), p. 1633-1651.
Lane, P.J., Lubatkin, M. (1998): Relative absorptive capacity and interorganizational learning. Strategic
Management Journal, 19 (5), p. 461-478.
Larsson, R., Bengtsson, L., Henriksson, K., Sparks, J. (1998): The interorganizational learning dilemma:
Collective knowledge development in strategic alliances. Organi:ation Science, 9 (3), p. 285-305.
Lettl, C. (2004): Die Rolle von Anwendern bei hochgradigen Innovationen. Eine explorative
Fallstudienanalvse in der Medi:intechnik. Wiesbaden: Dt. Univ.-Verlag.
Li, T., Calantone, R.J. (1998): The impact oI market knowledge competence on new product advantage:
Conceptual and empirical examination. Journal of Marketing, 62 (4), p. 13-29.
Lockett, F., Wright, M., Franklin, S. (2003): Technology transIer and universities` spin-out strategies. Small
Business Economics, 20 (2), p. 185-200.
Lynn, G.S, Akgn, A.E. (1998): Innovation strategies under uncertainty: A contingency approach Ior new
product development. Engineering Management Journal, 10 (3), p. 11-18.
Lynn, G.S., Morone, J.G., Paulson, A.S. (1996): Marketing and discontinuous innovation: The probe and
learn process. California Management Review, 38 (3), p. 8-37.
March, J.G. (1991): Exploration and exploitation in organizational learning. Organi:ation Science, 2 (1), p.
71-87.
Minniti, M., Bygrave, W. (2001): A dynamic model oI entrepreneurial learning. Entrepreneurship Theorv and
Practice, 25 (3), p. 5-16.
Narver, J.C., Slater, S.F. (1990): The EIIect oI a Market Orientation on Business ProIitability. Journal of
Marketing, 54 (4), p. 2035.
Nicholls-Nixon, C.L., Cooper, A.C., Woo, C.Y. (2000): Strategic experimentation: Understanding change and
perIormance in new ventures. Journal of Business Jenturing, 15 (5/6), p. 493-521.
Nonaka, I. (1994): A dynamic theory oI organizational knowledge. Organi:ation Science, 5 (1), p. 14-37.
Penrose, E. (1959): The theorv of the growth of the firm, OxIord: Blackwell.
Peters, M., Brush, C. (1996): Market inIormation scanning activities and growth in new ventures: A
comparison oI service and manuIacturing businesses. Journal of Business Research, 36 (1), p. 81-89.
Polanyi, M. (1958): Personal knowledge. Towards a post critical philosophv. Chicago, IL: University oI
Chicago Press.
Ravasi, D., Turati, C. (2005): Exploring entrepreneurial learning: A comparative study oI technology
development projects. Journal of Business Jenturing, 20 (1), p. 137-164.
Saemundsson, R.J. (2005): On the interaction between the growth process and the development oI technical
knowledge in young and growing technology-based Iirms. Technovation, 25 (3), p. 223-235.
Shane, S. (2000): Prior knowledge and the discovery oI entrepreneurial opportunities. Organi:ation Science,
11 (4), p. 448-469.
Sinkula, J.M. (1994): Market inIormation processing and organizational learning. Journal of Marketing, 58
(1), p. 35-44.
Slater, S.F., Narver, J.C. (1995): Does competitive environment moderate the market orientation-perIormance
relationship? Journal of Marketing, 59 (3), p. 46-55.
SteIIensen, M., Rogers, E.M., Speakman, K. (1999): Spin-oIIs Irom research centers at a research university.
Journal of Business Jenturing, 15 (1), p. 93-111.
Szulanski, G. (1996): Exploring internal stickiness: Impediments to the transIer oI best practice within the
Iirm. Strategic Management Journal, 17, Special Issue, p. 27-43.
Teece, D.J., Pisano, G., Shuen, A. (1997): Dynamic capabilities and strategic management. Strategic
Management Journal. 18 (7), p. 509-533.
Vohora, A., Wright, M., Lockett, A. (2004): Critical junctures in the development oI university high-tech
spinout companies. Research Policv, 33 (1), p. 147-176.
Von Hippel, E. (1986): Lead users: A source oI novel product concepts. Management Science, 32 (7), p. 791-
805.
Von Hippel, E., Tyre, M.J. (1995): How learning by doing is done: Problem identiIication in novel process
equipment. Research Policv, 24, p. 1-12.
Xie, W., White, S. (2004): Sequential learning in a Chinese spin-oII: the case oI Lenovo Group Limited. R&D
Management, 34 (4), p. 407-422.
Yli-Renko, H., Autio, E., Sapienza, H.J. (2001): Social capital, knowledge acquisition, and knowledge
Page 9 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
exploitation in young technology-based Iirms. Strategic Management Journal, 22 (6/7), p. 587-613.
Zahra, S.A., Ireland, R.D., Hitt, M.A. (2000): International expansion by new venture Iirms: International
diversity, mode oI market entry, technological learning, and perIormance. Academv of Management
Journal, 43 (5), p. 925-950.
Zahra, S.A., Sapienza, H.J., Davidsson, P. (2006): Entrepreneurship and dynamic capabilities: A review,
model and research agenda. Journal of Management Studies, 43 (4), p. 917-955.
Zucker, L.G., Darby, M.R., Armstrong, J.S. (2002): Commercializing knowledge: University science,
knowledge capture, and Iirm perIormance in biotechnology. Management Science, 48 (1), p. 149-170.
Page 10 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
Figure 1: Basic Research Model
H
1
H
3
Page 11 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
Table 1: Descriptive Statistics and Correlations of Main Variables
Table 2: Regression Results (Standardized Coefficients)
Variables Mean SD MP PL CL KI DYN
1. Market PerIormance (MP) 5.19 0.79 1.0
2. Personal Learning (PL) 4.01 1.20 .33 1.0
3. CodiIied Learning (CL) 3.81 1.15 .17 .46 1.0
4. Knowledge Integration (KI) 4.85 0.91 .34 .43 .31 1.0
5. Technological Dynamism
(DYN)
3.53 1.67 -.02 .16 .10 .09 1.0
Independent Variable Market PerIormance
Main effects.
Personal Learning (PL)
a
.20
t
CodiIied Learning (CL)
a
-.12
Knowledge Integration (KI)
a
.20*
Interaction effects.
PL x Technological Dynamism .20*
CL x Technological Dynamism -.25**
KI x Technological Dynamism -.58
Control variables.
b
Prior Knowledge
c
-.03
Existing Customers
c .40**
Founders Teamsize
c
-.27
t
Technological Uncertainty
c
-.16
t
Technological Dynamism
c
.09
Time Start-up to Launch -.02
R
2
(adjusted R
2
)
.53 (.40)
AR
2
step three
.06**
F 4.08**
N 71
** p _ .01; * p _ .05;
f
p _ .10 (one-tailed test oI coeIIicients)
a
Data was provided by second key inIormant (e.g. project manager)
b
Industry and technology Iield dummies not reported.
c
Data provided Ior the time oI start-up.
Page 12 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html
Figure 2: Interaction Plot of Personal Learning and Technological Dynamism
Figure 3: Interaction Plot of Codified Learning and Technological Dynamism
Page 13 oI 13
10/10/2011 http://Iusionmx.babson.edu/entrep/Ier/2007FER/cxxip2p.html