companies shared much of the same context in the early phases, VC ... Atlas Venture VC had invested in another 'data mining' company, so he knew the ...
Learning and venture capital: The growth of venture knowledge Henrik Berglund, Tomas Hellström, Sören Sjölander
Summary This paper sets out to analyze the modes of ‘venture learning’, which is defined as the process whereby knowledge is acquired on how to identify, develop and exploit business opportunities and the associated ventures. The context of learning discussed here is that of high technology ventures, and the ways in which certain types of venture relevant knowledge is acquired and transferred by venture capitalists to and from the venture process. Learning is modeled as two interrelated learning modes; hypothetico-deductive and hermeneutic learning, which turn out to be closely interrelated. The resulting model focuses on what is learnt, how it is learnt and how certain external actors can influence the quality, speed and cost of learning. Conclusions are made for research, policy and managerial practice with special focus on how learning can be effectively influenced by external actors such as venture capitalists. Keywords: Venture learning, venture capital, entrepreneurship, knowledge elicitation Introduction Every day, numerous budding entrepreneurs seek financial and managerial support from professional venture capitalist (VC). Very few receive such support, and even fewer go on to build successful companies. While much effort has been spent trying to understand the decision to invest and exit, considerably less attention has been paid to how VCs actually support entrepreneurs once the investment has been made, and the processes underpinning this support. A key point of departure is that new knowledge and new skills have to be acquired by inexperienced entrepreneurs and their teams, in the course of venture development. Given this implicit need, it is obvious that the process by witch such knowledge and skills are acquired that is ‘venture learning‘ - is fundamental to entrepreneurial success. One among several key issues here is how the venture as a process, involving one or several entrepreneurs, is interacting with others to create that competences needed to succeed. In this paper the issue of how VC expertise and VC knowledge elicitation can effectively influence productive venture learning is focused. Such an understanding may lead to better approaches to accelerated venture learning. In the course of this paper we will also illustrate how VC involvement may aid venture development, an issue that has received some previous, albeit scattered empirical attention (e.g., Busenitz et al. 2004). The aim of this paper then is to build a model grounded in learning theory, for how the VC acquires and elicits knowledge in relation to the entrepreneurial venture, and the ways in which
a VC may influence the acquisition of the needed new entrepreneurial competence. This process will be referred to as ‘venture learning’. In what follows we will present two illustrative cases of venture development and VC involvement, the process of which may be referred to as ‘VC driven venture learning’. This section will be followed by an account of the nature of VC expertise, which we divide into general and industry specific. Building on generic approaches to scientific development, we then discuss two modes of entrepreneurial learning, viz. hypothetico-deductive and hermeneutic. These learning modes can be found in abstract economic analyses of entrepreneurship, and also connected to specific contributions in the field of entrepreneurship studies. This account then paves the way for a discussion of how VCs can contribute to entrepreneurial learning and a model that accounts for the some of the key observations made in the cases below. Finally the theoretical discussion concludes with some policy suggestions and implications for VCs and entrepreneurs, as well as suggestions for future research. Two illustrative cases Two venture cases will help illustrate the influence of VC on the venture. While both the case companies shared much of the same context in the early phases, VC involvement differed greatly, as did the subsequent process of venture learning. Both companies were active in software, and were started by ambitious technology students from the same university at roughly the same period in time. Both received keen interest from the same seed investors, and while partly involving the same soft loan providers, they ended up being backed by different VCs, and with very different outcomes. Textile Solutions was started by two graduate students at Chalmers University of Technology in 1998. The aim of the business was to rationalize the apparel value chain and coordinate its market actors, from thread and cloth production to the store, in order to lower cost and increase market responsiveness through web based solutions. After a year of developing the concept and a prototype software solution, supported by former entrepreneur business angels and involving a few future potential customers, the business angels advised that VC money were needed to increase the scale of the technical effort, enhance sales capabilities and start selling to Swedish brands. The plan was to roll out on more markets after the concept had been verified in Sweden. Among the many VCs that signaled interest was InnovationsKapital with one managing partner in particular. Among others who had indicated strong interest were the VC firm Emerging Technologies with was headed by a former investment banker and PhD in economics. This VC was at that time one of the most well recognized VCs in Sweden, and he signaled very strong interest to invest. Wile InnovationsKapital was in the process of designing a term sheet and aligning interests with the Textile Solutions team, Emerging Technologies claimed that speed was key, and argued that ”you have to increase the scale! I don’t want to invest in a Shell Gas station, I want to invest in what will become Shell”. This VC demanded, as a prerequisite for the investment, that scaling up should be done promptly. He was ready to invest immediately and made the investment in Textile Solutions conditional on that the team rolled out their business not on one market (Sweden) but on eight markets simultaneously. His argument was that web based businesses depended particularly on speed of market penetration and first mover advantages. The young entrepreneurs
enthusiastically let go of InnovationsKapital, who was considered too slow, and instead opted for the aggressive Emerging Technologies as lead investor. The team of founders started to ambitiously implement the ‘enhanced business plan’ backed with money from several VCs, led by Emerging Technologies. However the basic concept did not gain market acceptance as fast as had been predicted and there were numerous technical challenges that needed serious rethinking and new solutions. After 20 month and 10 MUSD the company went bankrupt due to the fact that all the money had been used to build the wrong multinational organization for a technological concept that was not ready for launch, and which had not been verified commercially. With a revolutionary business model introduced into a conservative, socially well-structured market, uptake was very slow. At the same university only two years prior to Textile Solutions, in 1996, another student defended his thesis and started a company. The firm, ‘Spotfire’, was backed by InnovationsKapital, the VC firm later to be ‘rejected’ by Textile Solutions. Spotfire was intended to develop a type of data mining software, its algorithms invented by the founder, to allow users to graphically ’see correlations‘ between geometrical representations and structured numerical data. The applications was believed to have great potential in data mining of credit card user data, weather driven pollution data, and for production process quality data in the food and chemical industries, among others. In 1998 at the time of the founding of Textile Solutions, Spotfire was striving into its second year as a very small seed company trying to find its first real application and customers. The US market was seen as early adopters in this area, with numerous global brands, and with manufacturing firms and potential customers in food, finance, chemicals and the like. One dedicated InnovationsKapital VC, having prior experience selling quality monitoring solutions to US firms 3M and Mars Bars, advised the Spotfire team to try to bring in a US VC, in order to help in building the future organization on the US market. The team succeeded in bringing in AtlasVentures in Boston and one of their experienced investors. A few years earlier the Atlas Venture VC had invested in another ’data mining‘ company, so he knew the business and on the basis of these insights he advised Spotfire to develop a different application, and to use an alternative business model typical for similar software companies. The Atlas Venture VC also advised the Spotfire team to focus exclusively on one application – one that Spotfire themselves had never thought of - namely that of screening candidate molecules in preclinical research , that is finding the potentially fruitful molecule among hundreds of thousands of candidates. It was viewed as a productive means of finding out various properties (among others toxicity) of molecules by viewing their structures simultaneously as their physical and biological properties. Ten years later, in april 2006, the company is a ’visual interactive company‘ with all 24 of the biggest pharmas as repeat customers and with a range of other applications, it is ready for an IPO on NASDAQ or a trade sale. The view of the founder team is that this would very unlikely have happened without the good advice, practical assistance and hands on customer introduction, support for recruitment of key personnel and lastly access to down steam investors provided by Atlas Ventures and InnovationsKapital. In summary: While both start-ups had similar points of departure, VCs were differently involved yet decisively and distinctly influenced the outcomes, particularly through affecting the approach the entrepreneurial team were using when developing their original ideas. It is
clear that the Spotfire approach, much due to the VCs, was more explorative, incremental and also experimental, until the time they had found, through VC influence, their first core application. This can be compared to the approach of Textile Solutions which was led by the VC into planned albeit premature execution of a great market expansion, which ended up killing the company by burning their cash on the wrong business model. This business model and the assumptions about the dynamics of the market which underpinned it, was littered with hidden assumptions, some being fundamentally wrong. Textile Solutions never cleared out who was the customer for the product and service or if it was a product and/or a service, which value network participant should undersign the model, what the perceived and verified value was and how that value should have been appropriated, if network externalities were important and if there was a true first mover advantage. None of these assumptions were really questioned or in some cases even explicated. In the case of Spotfire however similar issues were gradually resolved by drawing on VC knowledge, either specific market knowledge or more general knowledge of what was needed to succeed, e.g. which geographical markets to enter, and what to do when. In what follows we will review the literature on VC expertise, focusing on specific industry expertise and general venture knowledge. The nature of VC expertise According to financial theory, the standard dept contract constitutes the most efficient form of financing under normal economic conditions (cf. Gale and Hellwig 1985). However, information asymmetries and the inherent uncertainty of entrepreneurship prevent the crafting of complete contracts that cover all future contingencies (Grossman and Hart 1986). Because of moral hazard, only knowledgeable investors who can monitor and influence the venture will invest. The moral hazard also goes in the other direction as ventures that depend heavily on the entrepreneur’s personal human capital is unlikely to accept external advice unaccompanied by a strong financial commitment. Financial theory thus argues that VCs must possess relevant expertise, especially when investing in uncertain and innovative ventures. Empirical studies of VC involvement reach similar conclusions. Surveys of both VCs and entrepreneurs consistently indicate that VCs are seen as most important in innovative early stage ventures, when the entrepreneur has limited experience, and when VCs hold a relatively large equity share. Together these results indicate that VCs need to possess relevant expertise to fulfill their role towards portfolio companies. However, the question remains as to what this expertise consists of. VCs expertise is often specialized, in that it concerns factors operating in a particular industry, and it is therefore typically based on a specific technological field and a limited market or even geographical area (Anderson 1999). In the first instance, experienced VCs may therefore possess technical expertise that can provide firms with guidance and direct assistance in developing and improving technical solutions. In the second instance the VCs market expertise can include the development of relevant business models, identification of relevant markets, actual marketing, as well as networking locally and across distances, thus contributing valuable new contacts associated with such activities (Sjögren and Zackrisson 2005).
In this way, the VC’s network complements that of the entrepreneur’s in important ways. VC networks are often a major source of unique market information and of candidates for employment on both high and low levels, as well as of service providers. VCs also help ventures promote themselves and get in contact with financiers, thus increasing their likelihood of securing additional funding (Gomez-Mejia et al. 1990). Apart from this, VCs also play a major role in strategy formulation, business modeling and implementation. This role may be particularly salient if the strategy hinges on unique offers and differentiation rather than on cost efficiency. This is because such strategies require knowledge of the needs of suppliers and buyers, of potential substitutes and of competing offers, knowledge that the VC is often more likely to possess. In a way these types of knowledge pertain to specific areas of relevance, which a VC becomes acquainted with after a certain time in a field of industry. However, beside such industry specific knowledge, experienced VCs have seen many ventures and have learned which strategies work and when. Competent VCs thus also possess a more general venture development expertise, which complements their industry specific knowledge, and which may be even more valuable, especially to inexperienced entrepreneurs. While such expertise may be specific to certain stages of investment (i.e. seed, early stage, expansion or buy out), one of the main selling propositions of VCs is their hard earned and generalizable expertise, when it comes to developing innovative ventures. This assumption, that such expertise may be present, is one which underlies much of the literature on organizational learning, and it suggests that VCs who have been involved in a range of ventures will bring their experiences to bear on subsequent decisions and activities (Busenitz et al. 2004). Experiences from both successes and failures will have imprinted into the VC a general understanding of what general principles for action, what constellation of people and resources etc., works and what doesn’t. Besides yielding better advice, an experienced VC can also provide comfort to less experienced CEOs and be a source of stability to the board of a new venture. This brings to mind the noninstrumental functions that VCs often come to play, for example that of mentors, confidants and friends, which are often brought out in interviews with entrepreneurs (MacMillan et al. 1989). Partly because of such tacit and informal aspects of the relation to the entrepreneur, experienced and successful VCs tend to transfer their positive image onto the ventures they work with. Another important contribution is that which could be called ‘knowledgeable discipline’. VCs know the importance of staying on course and try their best to pressure entrepreneurs to perform in accordance with jointly established objectives, when it is relevant to do so, or differently speaking, the import on persisting on the right track. If such managerial discipline is not maintained VCs will often replace the CEO. Generally speaking, VCs often know from experience how to handle complex interdependent organizational issues such as hiring and firing personnel, managing internal conflicts, and providing overall structure to emerging organizations. Hellmann and Puri (2002) investigated a large set of VC backed and non-VC backed ventures in the Silicon Valley region, and concluded that ventures obtaining venture capital were more likely to implement human resource policies, formal recruitment procedures, adopt stock option plans, replace CEOs, and recruit VPs for marketing and sales. These
findings indicate that VC firms influence their ventures in the process of professionalizing their operations, which is important not least on order to handle expansions. Two modes of venture learning It is apparent that experienced VCs have learned important lessons, which in different ways can help ventures develop. However in order to accurately examine how VCs contribute to such entrepreneurial learning we have to take a closer look at the modes of learning relating to venture development. Essentially, we will argue that mode and content of entrepreneurial learning are closely connected, an insight which has consequences for how to structure this relationship in time. Traditional economic theory tends to assume well informed agents who base their actions on rational analysis of given means-end frameworks. Action thus becomes choice, and choice merely calculation. As a result entrepreneurs are insignificant and entrepreneurial learning at best a marginal phenomenon (Baumol 1968, Bianchi and Henrekson 2005). In Austrian economics the market is instead seen as a process driven forward by alert entrepreneurs who continually discover new means-end frameworks. Some authors in this school have especially focused on entrepreneurship, but describes the entrepreneur qua economic ideal type to the detriment of a more practical understanding. In what follows we attempt a more practical understanding of entrepreneurial learning, by utilizing inroads to the problem provided by the work of two of Kirzner’s younger colleagues. The first of these, David Harper, describes how entrepreneurs learn by systematically testing and evaluating their business ideas in the marketplace. The second, Don Lavoie, instead tries to understand how entrepreneurs discover opportunities by emphasizing individuals’ cultural embeddedness. Both extend Kirzner’s notion of ’entrepreneurial alertness‘ and emphasize learning as the general guiding principle. They do obviously differ in terms of how such learning is to be translated into secondary principles, i.e. the actual process of entrepreneurial learning. These differences derive from the authors’ philosophical antecedents. Harper explicitly builds on Karl Popper’s hypothetico-deductive approach, in which learning is the outcome of discrete tests, which in turn emphasizes the context of justification, and where learning must be designed to live up to quite rigorous demands. Lavoie instead invokes HansGeorg Gadamer and the hermeneutic tradition, in which secondary principles are based on a gradually developing holistic preunderstanding of a phenomenon, and where search processes are more fluid, dialogical and difficult to formalize. Next Harper’s and Lavoie’s diverging views of entrepreneurial learning are described in some detail, and extended within the respective traditions in which they operate. This general discussion is related to complemented with more practical findings from the fields of management and entrepreneurship studies. Hypothetico-deductive learning David Harper (e.g. 1996) seeks to build a theory of entrepreneurial learning based on Karl Popper’s theory of scientific development. Harper delimits his analysis from the discovery of
new entrepreneurial opportunities. Following Popper he instead describes the logical procedure of how to enact them: “The Popperian programme takes the realm of logic and methodology (i. e. the context of justification) as the appropriate domain for philosophical analyses of the process by which knowledge grows. It does not take into account psychological, sociological and historical factors (i.e. the context of discovery).” (Harper, 1996: 31-32). The falsificationist entrepreneur, never takes his/her ‘theory’ to be ‘true’. Rather entrepreneurial assumptions are always of a speculative nature, like guesses or hypotheses, which may be overthrown in the light of opposing evidence or corroborated, i.e. be temporarily accepted (cf. Popper, 1963). Revisions are the key activity for people operating within such a tradition and Popper suggested the following model for how a discipline or a theory develops which can also be used to summarize the evolutionary, experimental mode of learning: • problem1 → tentative theory → falsification attempt or error elimination → refinement of theory (learning) → problem2.
This model suggest that learning starts from the encounter of a problem, in our context this could be an attempt to gain some arbitrage or capitalize a new scientific result. A set of hypotheses on how to solve this problem are then constructed. The entrepreneur puts these hypotheses to the test in the entrepreneurial situation, by actually trying out a business model, testing ideas on presumptive customers etc. What is left after these ‘tests’ have been performed is a new version of the problem, or an altogether new theoretical framework, should the original ideas not have panned out at all. Harper (1996) takes the entrepreneurial decisions to be sequential, quite in line with what Popper would have prescribed for science. Harpers model thus rejects the notion of long term planning in entrepreneurship. A number of management writers have developed more hands on suggestions for entrepreneurial management. Like Popper and Harper, these authors emphasize rigorous planning, testing and revising of hypotheses. This differs from planning in traditional business projects where deviations from the plan are seen as bad. New ventures, it has been argued, should instead design for systematic failure and embrace deviations as opportunities for learning (Sitkin 1992). This does not imply that demands on method and rigor are relaxed. Clearly in the Popperian tradition, some authors suggest that entrepreneurs ought to structure the gradual development of their ventures using four discovery driven planning documents, themselves akin to compound hypotheses. First, entrepreneur should create a reverse income statement. Backtracking from required (read hypothetical) profits, margins and sales volumes, entrepreneurs may roughly estimate the potential value of venture success. Second the venture idea is broken down into pro forma operations specs, which identifies underlying assumptions and, as far as possible, tests them in theory. Initially this can be done using a few phone calls, web searches etc. If the idea holds up under initial scrutiny, more detailed analyses are undertaken. In this manner, the entrepreneur can develop a reasonable estimate of the venture and assess the order of magnitude of its challenges. Third, a key assumptions checklist is developed. This is important to ensure that the critical assumptions that the venture is based on (as hypothesized in the pro forma operations specs) are flagged, discussed and checked as the venture unfolds. Finally, the assumptions are operationalized in a milestone planning chart which specifies how and in what order critical assumptions should be tested. These planning
documents are used to structure entrepreneurial learning and should be seen as a collection of assumptions which are continually revised as tests unfold. Furthermore, Sull (2004) proposes an explicitly Popperian approach to business development that consists of three sequential steps. First the entrepreneur should formulate a working hypothesis of what the opportunity is, what resources are required, what value the business may create, and how this value may be realized. It is especially important to identify potential deal killers and key drivers of success. Since assumptions often prove wrong, it is also important to keep the working hypothesis fluid, especially early on. Next, the entrepreneur should assemble the resources necessary to conduct experiments. This step is not part of the traditional Popperian framework but is critical to entrepreneurship. As a general rule, entrepreneurs should only assemble enough capital to conduct the most critical experiments. This is because too much capital may affect discipline and lead to sloppy behavior, e.g. neglect of potential deal killers. Finally, entrepreneurs should design and run experiments. Experiments in this context are tests designed to reduce sources of uncertainty that are critical to the success of the new venture. Typical experiments include customer research, building prototypes, limited introductions, tracking competitive reactions, and working with beta customers. Sull (2004) also discusses the relative merits of partial experiments, which address single sources of uncertainty, and holistic experiments, which are suitable when there are multiple interrelated sources of uncertainty. Entrepreneurs should also plan tests wisely, starting with low-cost partial experiments that target deal killers and key drivers of success. Once this is done, the entrepreneur may advance to more elaborate and expensive experiments. In the hypothetico-deductive tradition the context of discovery is only briefly touched on, before the authors lay out the “normative set of rules” (Harper 1996) that should guide the testing and evaluation of the discovered ideas. Sull (2004) states that the entrepreneur “begins the process by formulating a hypothesis” (Sull 2004: 72) but explicitly leaves for others to elaborate “the creativity to envision new things” (Sull 2004: 76). As acknowledged by both Popper and Harper, the context of discovery demands social and psychological insights and thus falls outside the realm of falsificationism (Popper 1963, Harper 1996: 31-32). It is however a fundamental aspect of real life entrepreneurship. To address this aspect of entrepreneurial learning, we therefore turn to hermeneutic perspectives of learning. Hermeneutic learning Harper assumed that the entrepreneurial context of justification, which is his focus, is independent from the entrepreneurial context of discovery (Harper 1996: 32). Lavoie instead focused on the context of discovery, and sought to better understand why entrepreneurs notice some but not other opportunities (cf. Lavoie 1990, Boettke et al. 2004). Lavoie argued that when entrepreneurs notice opportunities, which is required before any kind of hypothesis may be formulated, they do this from within their set of background assumptions and preunderstanding, and that these form the basis for any understanding of the world. Despite their intractability, the components and processes of such preunderstanding can be structured and described, albeit roughly and tentatively; a task taken on by sociologists and philosophers
such as Hans-Georg Gadamer, Alfred Schutz and Peter Berger, the last two to whom we will return shortly. From this perspective, entrepreneurial discovery and learning is not equal to intractable alertness, followed by a rational process of weeding out faulty hypotheses. Instead, it rests on structured yet often tacit preunderstanding and a process whereby entrepreneurs with different experiences try to make sense of and influence the world around them: “The seeing of profit opportunities is a matter of cultural interpretation. And like any other interpretation, this reading of profit opportunities necessarily takes place within a larger context of meaning, against a background of discursive practices, a culture.” (Lavoie 1991a: 36) The hermeneutic approach to learning encompasses two general aspects. First, the preunderstanding of the actors involved, or alternatively their network of more or less tacitly held assumptions about the world. Second, the process by which this preunderstanding is transformed. In the first case, one may fruitfully draw on Alfred Schutz’ and Thomas Luckmann’s concept of a ‘stock of knowledge’, developed by them in The Structures of the Life World (1973). Schutz and Luckmann argued that the stock of knowledge of a person or a collective was built up over time and consisted of various sets of cognitive and emotional pointers (alternatively interests or relevancies). These are topical and focuses a person’s attention on certain themes, interpretative and confer meaning vis-à-vis experiences and objects, and finally motivational, namely when and how the actor is stimulated into action. The notion of ‘stock of knowledge’ is a good starting point for grasping in what dimensions of understanding hermeneutic learning takes place. In addition to ‘relevancies’ a person’s stock of knowledge is, according to Schutz and Luckmann, made up of types and categories, more concrete representational and often possibly explicable representations of objects and relations in the world, for example of hierarchies of things. The build-up of a stock of knowledge or understanding has been suggested to take place as a circular movement between the concrete parts of something (e.g. a particular factual insight or utterance) and the totality of meaning within which interpretation takes place (cf. Gadamer’s ‘hermeneutic circle’, Gadamer 1976). In line with this Brown and Duguid argue that central to both innovation and learning is the development of a “new way of seeing” or a new interpretive view (Brown and Duguid 1991). Counter to the Popperian assumption that the entrepreneur first sets up and then tests abstract hypotheses, this tradition argues that action may precede cognition, and that learning and sense-making occurs when “a flow of organizational circumstances is turned into words and salient categories” (Weick et al. 2005: 409). This form of learning can be said to be especially relevant in entrepreneurial contexts, where often very little of consequence is prestructured in abstract propositional form. Gartner for example argues that emerging organizations are characterized by ambiguity, and that the role of the entrepreneur is to gradually overcome this ambiguity by developing ever more plausible interpretations (e.g. Gartner et al. 1992, 2003) of the future. Learning is thus not a matter of systematically discovering the true nature of the world, but of arriving at more reasoned and useful interpretations: “there is no such thing as a representation that is true or false, there are simply versions that are more or less reasonable” (Weick 1979: 169, quoted in Gartner et al. 2003). These interpretations are based on the
entrepreneur’s prior knowledge and experiences (emerging preunderstanding or stock of knowledge), including the experiences of their social networks. Since entrepreneurship is a practical activity, mere interpretations of the past are not enough. The interpretations are important because they allow the entrepreneur to envision and “act as if” the future is given, which in turn provides new grist for learning and sense-making (Gartner et al. 1992). An important part of this logic is thus that acting precedes thinking. Entrepreneurs do not elaborate hypotheses which are then tested. Instead actions lead to refined interpretations through an iterative process of enactment and sense-making, directed by and adding to a growing stock of knowledge. Venture learning it thus a matter of acting with other people in a way that gradually develops the entrepreneurial resource base and interpretive framework (cf. Alvarez and Busenitz 2001). Table 1. Summary of important differences between hypothetico-deductive and hermeneutic views of learning Hypothetico-deductive Falsification experimentation
of
Hermeneutic hypotheses, Elaboration of understanding
interrelated
forms
of
Learning is developed by individuals and Learning is developed in collectives but shared to collectives articulated by individuals Learning is structured with focus on well Learning is emergent with unknowable defined outcomes outcomes Cognition precedes and provides reasons People act on the basis of their tacit for learning experiments. Preunderstanding preunderstanding. Learning is refinement not relevant of this preunderstanding Mainly focused on the context justification, and finalized pieces knowledge
of Mainly focused on the ‘ongoing’ and of unfolding context of discovery
Discussion and implications The previous discussion has emphasized venture learning, and the VCs role in this process, as an external elicitor of new and pertinent knowledge bearing on the venture process. In this capacity the VC has been described as an expert who possesses both general venture knowledge and industry specific knowledge relevant to the venture and its entrepreneurs. It has also been argued that similar to knowledge creation in science, entrepreneurial and venture learning can be expressed in terms of two main modes, namely hypothetico-deductive or experimental, with short feedback loops and incremental, factual updating of knowledge, and hermeneutic, with changes in understanding of more fundamental aspects of the field of
activity. As seen above, these modes have been successfully applied to science as well as to entrepreneurship, and it is therefore likely that they also apply to the VCs role as knowledge elicitor in the venture process. Drawing on these previous arguments it is now possible to suggest a synthesis model of venture learning and the way VCs support this process. This model will be explicated together with a number of normative implications for the role of the VC in venture learning. Figure 1 models the knowledge elicitation of the VC through time, and demonstrates the mechanisms of the resulting venture learning, as has been outlined above. The two horizontal arrows of the model represent knowledge build in the two main modes discussed above: ‘Acting’ has been added to the hypothetico-deductive mode to emphasize that this is a more experimental, testing approach to knowledge creation, in the sense that creating new knowledge in this sphere involves collecting new data. ‘Reflecting’ has similarly been added to the hermeneutic stream of knowledge buildup to emphasize the predominantly reflective or speculative approach of this mode. In addition we see four types of within and between transfer relating to these to modes, which together make up the VCs relation to venture learning. We will now go through these in turn, and relate them to the previous discussion and initial case descriptions. Figure 1. A synthesis model of venture learning 3
Acting (H-D) 1
2
Reflecting (H) 4
1) Experimentation Based on a given preunderstanding anchored in prior experiences, the VC may formulate and subsequently test a hypothesis in relation to the venture. This hypothesis aims at confirming or rejecting an assumption about technology, market, business model, strategy, etc. General criteria when formulating such hypotheses could be: a) They must not be too complex to implement and evaluate b) They must not be too expensive to implement and evaluate c) They must not take too long to implement and evaluate From a normative perspective we note that VCs should attempt ‘optimize’ learning given these criteria. For example, Sull (2004) takes up the differences between partial and holistic experiments, and methodology handbooks describe various forms of experiments (critical case, maximum variation, extreme case, etc.). It is clear from the two cases above that VCs may be decisively influential during this phase of venture learning, in the sense that they can contribute
greatly, as well as inflict a lot of damage. In the case of Textile Solutions for example we observed how the VC’s basic appreciation of competitive pressures in the market led to overexpansion and bankruptcy, while Spotfire was encouraged to focus on one application in order to reduce complexity. The key role for the VC in this phase is to surface hidden assumptions in the venture, to identify critical uncertainties and to help device relevant hypotheses that address critical issues in the venture. 2) Reflection After a hypothesis has been tested in action, the results may be reflected upon and incorporated into a new preunderstanding or ‘worldview’. This new preunderstanding is used to reconfigure the venture and it also forms the basis for additional experiments. Here VCs fill an important role as ‘decoders’ of the outcomes of tests, as these have to be related to an understanding of the overall business situation. In this process the VC may help entrepreneurs make sense of the outcome of a particular action in terms of the importance to the stock of knowledge previously acquired by the VC, either in the sense of general venture expertise or in terms of their industry specific understanding. The VC may be especially critical in terms of evaluating unexpected outcomes, e.g. strange customer responses and suggestions (cf. Sull 2005). This process of reflection may end up in suggestions to reconfigure the business model, pursue other target customers, focus technology development in a different direction, seek out certain strategic partners, etc. The case of Spotfire is especially illustrative in this regard. Here we observed how they drew on the initial response from the pharmaceutical industry to reconfigure themselves as a ‘visual interactive company’, which in turn spawned new hypotheses regarding product lines, customers and business models. Reflection and reconfiguration of the venture will in turn generate new hypotheses, thereby creating a movement between experimenting and reflecting. 3) ‘Unreflected’ action Sometimes VCs and entrepreneurs seem to act without a clear and explicit purpose, for example as a result of routine or ‘standard operating procedures’. Following the arguments above this is potentially wasteful, since while actions, planned or otherwise, do not always target issues that are important to the venture; a reasoned action has a higher likelihood to generate learning than an unreasoned one. However, sometimes unreflected/unplanned actions can stumble across important results, precisely because they are not consciously planned. In this process the role of the VC may be to help interpret the results of unplanned actions and to help entrepreneurs compensate for mistakes. It is likely that this type of action must be tolerated to some extent, if not for any other reason because it is unavoidable. However, care should be taken not to allow too much unreflected action. This was demonstrated in the case of Textile Solutions which started implementing a poorly thought through business model, which did not function very well with the conservative textile industry, according to a logic that resembled routine. 4) Unverified assumptions.
The final arrow depicts how VCs and entrepreneurs may jointly develop novel assumptions without really exposing their ideas to real life tests. Unverified assumptions may be both explicit and implicit. In the first instance the VC/entrepreneur essentially takes a calculated risk by chancing that an assumption is correct. If instead assumptions are implicit, the entrepreneur may be said to be ‘unconsciously incompetent”, something which may lead to disastrous results. The Spotfire case illustrates the positive alternative to this mode, namely a clear exchange between assumptions and real world tests. A remedy to the first instance might be for the VC/entrepreneur to develop a ‘central hypothesis checklist’ and decide when to take a chance and when to conduct tests in action. Here it is clear that the competent VC is extremely important, for similar reasons as he/she is a key resource in the formulation of hypotheses to test. Conclusions It is obvious that VCs do not simply rely on effective selection and monitoring of ventures to deliver premium performance. Effective guidance during development of the venture is also pertinent, and this in turn depends on venture learning – a process in which the VC plays a central role. In the above we have reviewed a number of such learning processes as they pertain to various aspects of the venture. We have argued that the VC and the entrepreneur form a learning alliance where industry specific and general venture knowledge is exchanged, and where the VC by dint of previous experience and oversight has a key role to fill in this exchange. We have further argued that knowledge is built or elicited in this process by means of incremental additions to propositional knowledge as well as by reflection and a more systemic reformulation of the foundations for acting in the world. Following received conceptual canons, we have referred to these modes as hypothetico-deductive and hermeneutic respectively. The process of positive venture learning has consequently been conceptualized as an exchange between these two modes, where one drives the other. Here, the hypotheticodeductive mode may also be referred to as ‘empirical’ while the hermeneutic can be seen as a form of ‘theory building’ activity. This helps us understand venture learning as a species of learning in general, including scientific learning and ‘biological learning’, both of which are often viewed as a form of blind variation and retentive selection, where the effects of phenotype ‘experiments’ are stored in the genetic memory of a lineage (Andersson 1999, Hull, 1993). We have also identified two potential dysfunctional forms of venture learning viz ‘unreflected action’ and ‘unverified assumptions’, which would prevent such learning from taking place. Given previous discussions, one important issue that emerges is how, once the investment has been done, the skill of the VC practically can help accelerate entrepreneurial learning and positively affect the value of the company. As we have seen from the cases above, the ability to create relevant experience-based hypotheses concerning applications, customers, benefits and customer value, development processes, sourcing structures, distribution channels, sales organization, revenue models, financial setup etc are key, as is the working methodology of ‘exploring while exploiting’. In this process the VC can make a real difference in adding value to a venture. However, in order to increase the ability of the VC to add value to the venture he or she not only has to define his/her role as one of choice-maker, selector and provider of capital, but must more importantly see him/herself as a provider of key competence and a
steward and guide in the process of creating and eliciting knowledge about the commercial value of new ideas held by entrepreneurs. The present paper should be seen as a contribution towards understanding the anatomy of the process by which such transformation may be achieved and developed. References Alvarez, S., Busenitz, L. (2001). The entrepreneurship of resource based theory. Journal of Management 27: 755–776. Anderson, P. 1999. Venture capital dynamics and the creation of variation through entrepreneurship. In Baum, J.A.C. and McKelvey, B. (eds) Variations in Organization Science: In Honor of Donald T. Campbell. Thousand Oaks, CA: Sage. Baumol, W. (1968). Entrepreneurship in economic theory. American Economic Review, 58(2): 64-71. Bianchi, M. and Henrekson, M. (2005). Is neoclassical economics still entrepreneurless?. Kyklos. 58(3): 353-377. Boettke, P., Lavoie, D. and Storr, V. (2004). The subjectivist methodology of Austrian economics and Dewey’s theory of inquiry. In Khalil, E. (ed.), Dewey, Pragmatism and Economic Methodology. London: Routledge, pp. 327-356. Brown, J. and Duguid, P. (1991). Organizational learning and communities of practice: Toward a unified view of working, learning, and innovation. Organization Science. 2(1): 40-57. Busenitz, L. W., Fiet, J. O. and Moesel, D. D. (2004) Reconsidering the Venture Capitalists' "value added" Proposition: An Interorganizational Learning Perspective. Journal of Business Venturing 19(6):787-807 Casson, M. (2005). The Entrepreneur: An economic theory, Edward Elgar, London, 2005 Fiet, J. (1996). The informational basis of entrepreneurial discovery. Small Business Economics. 8: 419-430. Gadamer, H. (1976). Philosophical Hermeneutics.Berkeley: University of California Press. Gale and Hellwig (1985). Incentive-Compatible Debt Contracts: The One-Period Problem. Review of Economic Studies. 4: 647-633 Gartner, W. Bird, N. and Starr, J. (1992). Acting as if: Differentiating entrepreneurial from organizational behavior. Entrepreneurship: Theory & Practice. 16(3): 13-31. Gartner, W., Carter, N. and Hills, G. (2003). The Language of Opportunity. In Steyaert, C & Hjorth, D. (Eds.) New Movements in Entrepreneurship. London: Edward Elgar, pp. 103-124. Gifford, S. (2003). Risk and uncertainty. In Acs, Z. and Audretsch, D. (eds.), The Handbook of Entrepreneurship Research. New York: Kluwer, pp. 37-54 Gomez-Mejia, L. R., Balkin, D. B. & Welbourne, T. M. (1990). Influence on venture capitalists on high tech management. The Journal of High Technology Management Research 1: 103-118. Grossman, S. and Hart, O. (1986). The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration. Journal of Political Economy. 94(4): 691-719. Harper D. (1996) Entrepreneurship and the Market Process, London, Routledge Hellmann, T. and Puri, M. (2002). Venture Capital and the Professionalization of Start-ups: Empirical Evidence. Journal of Finance. 57(1): 169-97.
Hull, D. (1988). Science as a Process. Chicago University Press, Chicago. Lavoie, D. (1986). The Market as a Procedure for Discovery and Conveyance of Inarticulate Knowledge. Comparative Economic Studies. 28. Lavoie, Don 1991. “The Discovery and Interpretation of Profit Opportunities: Culture and the Kirznerian Entrepreneur.” In Brigitte Berger, ed. The Culture of Entrepreneurship. San Francisco: Institute for Contemporary Studies. Lavoie, Don. ed. 1990. Economics and Hermeneutics. London, UK: Routledge. LR Gomez-Mejia, DB Balkin, TM Welbourne. Influence of venture capitalists on high tech management Journal of High Technology Management Research, 1990 MacMillan et al. 1989 JBV March (1982) J.G. March, The technology of foolishness. In: J.G. March and J.P. Olsen, Editors, Ambiguity and choice in organizations, Universitetsforlaget, Bergen, Norway (1982), pp. 69–81. McGrath, R.G. (1999) Falling forward: real options reasoning and entrepreneurial failure. Academy of Management Review, Vol. 24(1), pp 13-30 MacMillan, I. C., Kulow, D. M. & Khoylian, R. (1989) Venture capitalists' involvement in their investments: Extent and performance. Journal of Business Venturing 4: 27-47. Minniti and Bygrave 2001 "A Dynamic Model of Entrepreneurial Learning. Entrepreneurship: Theory and Practice. 25(3) 5-16. Popper, K.R. (1963). Conjectures and Refutations. Routledge Kegan and Paul, London. Porter, M. (1996). What is Strategy?. Harvard Business Review. 74(6): 61-78. Sarasvathy, S. (2004). Constructing corridors to economic primitives: Entrepreneurial opportunities as demand-side artifacts. In Butler, J. (ed.), Opportunity Identification and Entrepreneurial Behavior. Greenwich CN: Information Age Publishing, pp. 291312. Sarasvathy, S. and Dew, N. (2005). Entrepreneurial logics for a technology of foolishness. Scandinavian Journal of Management. 21(4): 385-406. Schutz, A. and Luckmann, T. (1973). The Structures of the Life-World. Northwestern University Press, Evanston. Simon, H. (1996). The Sciences of the Artificial. Third edition. Cambridge, MA, MIT Press. Sitkin, S. (1992). Learning through failure: the strategy of small losses. Research in Organizational Behavior 14: 231–266 Sjögren, H & Zackrisson, M, 2005, “Financing of high technology small firms in Sweden and USA”, Venture Capital, vol. 7, No 1, 23 p. 30 Spinosa, C., Flores, F. and Dreyfus, H. (1997). Disclosing New Worlds: Entrepreneurship, Democratic Action, and the Cultivation of Solidarity. Cambridge, MA: The MIT Press. Weick, K. (1979). The Social Psychology of Organizing. New York, NY: McGraw Hill. Weick, K., Sutcliffe, K. and Obstfeld, D. (2005). Organizing and the Process of Sensemaking. Organization Science. 16(4): 409-421.
