FISH OUT OF WATER: TRANSLATION, LEGITIMATION, AND NEW VENTURE CREATION Paul Tracey Cambridge Judge Business School University of Cambridge Trumpington Street Cambridge CB2 1AG Tel: +44 (0)1223 760471 Email: p.tracey@jbs.cam.ac.uk  Elena Dalpiaz Imperial College Business School Imperial College London South Kensington Campus London SW7 2AZ Tel: +44 (0)20 7594 1969 Email: e.dalpiaz@imperial.ac.uk Nelson Phillips Imperial College Business School Imperial College London South Kensington Campus London SW7 2AZ Tel: +44 (0)20 7594 1699 Email: n.phillips@imperial.ac.uk Acknowledgements: We are indebted to our associate editor and three anonymous reviewers who made such a significant contribution to the development of our work. We also thank participants of the Oxford Reputation symposium (2017), OTREG meeting (2017), EGOS colloquium (2015), AOM conference (2014), and ASQ/OMT conference (2013), as well as seminar participants at WU, Rotterdam School of Management, Newcastle University Business School, and Warwick Business School, who offered many valuable insights and suggestions. FISH OUT OF WATER: TRANSLATION, LEGITIMATION, AND NEW VENTURE CREATION ABSTRACT We draw on institutional theory to study a common type of new venture creation that has been neglected in the literature: the translation of an existing organizational form from a different – and misaligned – institutional context. To do so we conducted an in-depth case study of H-Farm, an Italian venture that was founded as a business incubator, a type of organization that first emerged in Silicon Valley and other US technology regions. Our study illuminates the specific configuration of legitimacy pressures inherent in this type of entrepreneurship, and theorizes the strategies that entrepreneurs can enact to address them: local authentication work, category authentication work, and dual optimal distinctiveness work. We also show that the legitimacy pressures experienced by entrepreneurs may vary significantly as ventures mature, and challenge the notion of a specific “legitimacy threshold” that new ventures are required to reach. Finally, our model conceptualizes translation as an iterative, dynamic and ongoing accomplishment rather than a “one off” activity with clear beginning and end points. New ventures are often created through the translation of an existing organizational form from a different institutional setting, even when that form is a poor fit with the local institutional context of the new venture. An organizational form is an archetypal configuration of goals, practices and identity claims that mark members out as belonging to a particular category of organizations (Greenwood & Suddaby, 2006; Tracey, Phillips, & Jarvis, 2011; Vergne & Wry, 2014). Examples of new ventures created in this way include Starbucks, which translated an organizational form – the specialty coffee retailer – from Italy to the US at a time when Americans’ taken-for-granted understanding of coffee shops was places offering poor quality coffee to go (Rindova & Fombrun, 2001). Similarly, Ryanair and EasyJet translated Southwest Airline’s model to Europe at a time when the core features of low-cost air transportation did not fit European understandings of what an airline should look like and what services it should offer (Rae, 2001). Despite the prevalence of the phenomenon, our understanding of this kind of new venture creation is limited. Yet it poses a distinct and theoretically compelling set of questions. From an institutional perspective, the questions with respect to legitimacy are especially intriguing (Fisher, Kotha, & Lahiri, 2016; O’Neil & Ucbasaran, 2016; Tolbert, David, & Sine, 2011). Specifically, entrepreneurs seeking to translate an organizational form from a misaligned institutional context – i.e., a context characterized by different expectations about ‘appropriate’ organizational behavior (Scott, 1995) – are presented with a complex set of legitimacy pressures that remain largely unexplored. First, the translated venture needs local-level legitimacy – i.e., to be viewed as a legitimate organization in the institutional context in which it is created – so that it gains acceptance among local-level stakeholders and access to their resources (e.g., Kostova, 1999; Rao, Morrill, & Zald, 2000). Second, in order to grow, the translated venture is likely to need category-level legitimacy – i.e., to be viewed as a legitimate member of the organizational form on which it is based – so that it gains acceptance among category-level stakeholders, which are located in the institutional context where the form originated, and access to their resources (e.g., Zuckerman, 1999; Vergne & Wry, 2014). Third, as the translated venture matures, legitimacy can have negative, as well as positive, consequences. Specifically, conformity to institutional expectations may make it harder for a given venture to distinguish itself from rivals in the eyes of local- and category-level stakeholders. Thus, new ventures face the challenge of becoming “optimally distinct” (Zhao et al., 2017) – legitimate but different – vis-à-vis other members of the same form at both the local- and category-levels to secure continued access to resources. To understand how entrepreneurs experience and manage these pressures throughout the translation process, we conducted a qualitative case study of H-Farm, an Italian venture founded in 2005 that combines seed incubation, venture capital investment, education, and consulting in the digital, web and new media industries. The organizational form that H-Farm’s founders used as a target model was the business incubator, a category of organizations that they understood as originating in Silicon Valley, and which is itself designed to support entrepreneurial ventures (Kenney, 2000). Based on our findings, we develop a theoretical model of ‘entrepreneurship through misaligned translation’. Our model highlights the complex pressures at the local- and category-levels that entrepreneurs engaged in this kind of translation must navigate. To respond to these pressures, entrepreneurs enact different forms of work: local-level authentication work, category-level authentication work, and dual optimal distinctiveness work. Interestingly, we find that these types of work unfold over three discrete translation phases – improvising, converging, and optimizing – as the target form being translated, as well as the nature of the local- and category-level legitimacy pressures, evolve. We further show that these kinds of work can become intertwined, generative processes that allow ventures to create a distinctive way of being a member of a given form. In developing our arguments, we make three contributions to the institutional perspective on entrepreneurship. First, we contribute to the growing literature on legitimation and new venture creation. Specifically, we illuminate the particular configuration of legitimacy pressures inherent in the creation of a new venture through the translation of an organizational form from a misaligned context, and uncover the entrepreneurial strategies required to address it. Second, we shed light on the temporal dynamics of new venture legitimation. In doing so we challenge the idea that new ventures need to achieve a specific “legitimacy threshold” (Zimmerman & Zeitz, 2002) to survive and grow. Finally, we contribute to institutional research on translation by showing that, in the context of entrepreneurship, translation is a highly uncertain process in which precise emulation is unlikely to be effective or even feasible. LEGITIMACY, TRANSLATION, AND NEW VENTURE CREATION Attaining organizational legitimacy – “a generalized perception or assumption that the actions of an entity are desirable, proper or appropriate within some socially constructed system of norms, values, beliefs and definition” (Suchman, 1995: 574) is fundamental for new venture survival and growth. Nascent ventures face a liability of newness (Stinchcombe, 1965). Specifically, the lack of a track record means that key stakeholders in their institutional environment are often reluctant to endorse them, and may even question the reason for their existence, with the risk that they are starved of the resources required to thrive (Fisher et al., 2016). To become legitimate is to garner the approval of these key institutional stakeholders, and the performance of nascent ventures is strongly linked to entrepreneurs’ effectiveness in doing so (Aldrich & Fiol, 1994). Researchers have devoted increasing attention to how entrepreneurs overcome their venture’s liability of newness to become legitimate (O’Neil & Ucbasaran, 2016; Überbacher, 2014). This work has shown that entrepreneurs engage in different types of action to convince resources providers that they are ‘worthy’ of support (Tolbert et al., 2011). For example, they rely on “practice work” (Gawer & Phillips, 2013) to design a set of practices that allows the venture to operate effectively in a given institutional context (Tracey et al., 2011). They also rely on “meaning work” (Rao & Giorgi, 2006) – the strategic use of symbols (Zott & Huy, 2007), stories (Lounsbury & Glynn, 2001), and metaphors (Cornelissen & Clarke, 2010) – to construct shared understanding about these practices so that they ‘make sense’ in that context. The aim is to “shape the attention and perceptions” (Wry, Lounsbury, & Glynn, 2011: 450) of key stakeholders in order to win their backing for the emerging organization. Such actions are important, because legitimacy requires that new ventures conform to institutionalized prescriptions about firm behavior – stakeholders will penalize ventures that deviate too far from expectations (Zimmerman & Zeitz, 2001). This literature has made important advances in understanding how new ventures attain legitimacy. Interestingly, however, it has focused mainly on the struggles for legitimacy in a single institutional setting, usually where the ventures in question belong to a recognized organizational form (e.g., Clarke, 2011; Cornelissen, Clarke & Cienki, 2012; Stone & Brush, 1996; Zott & Huy, 2007). By contrast, the actions involved in translating a venture that belongs to an organizational form that is legitimate in another institutional context, but not in the context where it is being set up, has been studied much less systematically. Yet, the legitimacy dynamics in these circumstances are distinct. Specifically, entrepreneurs are faced with two discrete sets of legitimacy pressures: 1) the need for local-level legitimacy among stakeholders located in the institutional context in which the venture was founded, and 2) the need for category-level legitimacy among stakeholders of the target form, that are located in the institutional context from which the form was translated (cf., Voronov, De Clercq & Hinings, 2013). Stakeholders in both contexts include, for example, investors, customers, employees, and peers. A translated venture with local-level legitimacy is considered “appropriate” and “desirable” (Suchman, 1995) among stakeholders in the context in which it was created, thereby unlocking access to their resources (Gardberg & Fombrun, 2006; Xu & Shenkar, 2002). This requires that the venture conforms to local conceptions about “what type of actors are allowed to exist, what structural features they exhibit, what procedures they can follow, and what meanings are associated with these actions” (Ruef & Scott, 1998: 879). Local legitimacy is critical for a new venture. Without it, an organization cannot grow the roots needed to establish and sustain itself – it is starved of the requisite resources to survive (Kostova, 1999). For translated ventures that are content to remain small or whose local contexts contain abundant resources, local-level legitimacy may be sufficient. As soon as these ventures start to compete for resources outside their locale, however, the pressures for legitimacy at the category-level increase. Entrepreneurs who translate a form to very large market settings, such as the US or China, may be more likely to find the requisite resources locally than entrepreneurs who translate a form to a European country, whose local contexts are constrained by their relatively small scale. However, even in the largest market settings, many translated ventures need to look beyond their local contexts “to acquire strategic resources and reduce their institutional and market constraints at home” (Luo & Tung, 2007: 481) and therefore need category-level legitimacy, as illustrated by the example of Chinese ride-hailing firm Didi, described in the next section. A translated venture with category-level legitimacy is considered “appropriate” and “desirable” (Suchman, 1995) among stakeholders of the target form, thereby unlocking access to their resources (Navis & Glynn, 2010). These stakeholders are inevitably located in another institutional context. Failure to attain category-level legitimacy is likely to inhibit growth as resource providers outside a venture’s local context tend to discount organizations that do not meet “institutionalized expectations for how they should look and act” (Zuckerman, 1999: 1399; Vergne & Wry, 2014). However, category-level expectations are not fixed: the meanings associated with category membership and the boundaries of the category itself are continually negotiated through social interactions (Khaire & Wadhani, 2010; Jones, Maoret, Massa, & Svejenova, 2011; Granqvist & Ritvala, 2016). These interactions are shaped by the social context in which they take place and the goals and interests of the actors involved (Grodal & Kahl, 2017). In conceptualizing category-level legitimacy in this way, we align with a growing body of work that challenges the “conventional simplification” of categories as ideal prototypes, and instead views categorization as a “socially, politically, and temporally situated, dynamic process” (Durand, Granqvist, & Tyllestrom, 2017: 5). Distinguishing Types of New Venture Creation Through Translation There are different ways in which entrepreneurs can translate existing organizational forms to create new ventures, which influence profoundly the nature of the local- and category-level legitimacy dynamics. To help make sense of this variation, we distinguish two dimensions along which approaches to new venture creation through translation can be characterized. First, the type of new venture created can vary. The venture can be a part of an existing organization such as a subsidiary of a MNC (e.g., Li, Yang, & Yue, 2007) or of an entrepreneurial venture (e.g., Bingham, 2009); or the venture can be a de novo organization – an entrepreneurial venture with no ties to a parent organization. Second, the degree of alignment between institutional contexts can vary. Institutional alignment refers to the extent to which the expectations of ‘appropriate’ behavior differ between the context in which the organizational form being translated is embedded, and the context in which the new venture is being created (Scott, 1995; Ruef & Scott, 1998)[footnoteRef:1]. The two contexts can be closely aligned, or they can be misaligned. These distinctions are seldom explicitly articulated in the literature, but we think it is important to make them in order to be precise about the phenomenon that we are investigating (cf., David, Sine & Haveman, 2013). Table 1 uses these two dimensions to summarize a new venture’s initial legitimacy position in the eyes of local and category-level stakeholders for four different kinds of translation, and to locate our study in the literature. [1: The notion of institutional alignment resonates with the concepts of “institutional distance” and “cultural distance” in the internationalization literature (e.g. Kostova, 1999; Bingham, 2009). ] ----------------Insert Table 1 about here-------------- Quadrant 1 in the table represents the simplest form of translation – an existing organization that is recognized as part of an existing organizational form creating a venture in another institutional context, where that context is relatively closely aligned with its own. In these circumstances, category-level legitimacy has already been attained and local-level legitimacy should be straightforward because the institutional context of the new venture is similar to that of the existing venture. Uber, the pioneering digital ride-hailing firm founded in San Francisco in 2009, entering Canada is an example of this type of translation. Quadrant 2 represents a more difficult translation scenario – an existing organization that is recognized as part of an established organizational form creating a venture from another institutional context that is misaligned with its own. In these circumstances, category-level legitimacy is already in place but local-level legitimacy is likely to be problematic because the context of the new venture is very different to that of the target organizational form. Uber entering India in 2014 is an example of this type of translation. Quadrant 3 represents a different type of translation again – a de novo organization created in an institutional context that is closely aligned with the institutional context of the target organizational form. In these circumstances, local-level legitimacy should be relatively straightforward to achieve because the institutional context of the new venture is similar to the context where the form originated, but the venture is still tasked with establishing its legitimacy as a member of the target form. Hailo, the British technology firm that translated the digital ride hailing model from California to the UK in 2011, is an example of this type of translation. Quadrant 4 represents the most complex type of translation – a de novo organization founded in an institutional context misaligned from the context in which the target form is embedded. In these circumstances, both local-level and category-level legitimacy can be particularly challenging to attain: local-level legitimacy is likely to be problematic because the context of the new venture is very different to that of the target form, and the venture also faces the challenge of becoming legitimate among category-level stakeholders in a very different institutional context. Didi, the Chinese company that translated the digital ride hailing model from the US to China in 2012, is an example of this kind of translation. Interestingly, Didi initially struggled to attain local-level legitimacy because it operated in a grey legislative area and because the concept of digital ride-hailing contrasted with local understandings of how fares were calculated and paid for. Over time, Didi became more legitimate in China and local legitimacy pressures eased, as evidenced by the legalization of ride-sharing in 2016, and massive investment by Chinese internet giants including Alibaba in 2017 (Yang & Liu, 2017). Despite its increased local-level legitimacy and access to significant local capital, Didi also sought resources, and hence legitimacy, from category-level stakeholders, because the Chinese context was resource constrained in key respects. For example, Didi wanted to take a lead in the development of driverless cars, but the cutting-edge research infrastructure required to do so was not available in China but rather in California, the place where digital ride hailing originated. Didi was therefore exposed to legitimacy pressures from stakeholders in Silicon Valley because it needed to persuade resource providers in that context to support it. The venture did so successfully, raising $1bn from Apple in 2016 and opening Didi Labs in Mountain View in 2017, a purpose built facility for research into artificial intelligence technology for driverless cars (Chen & Elstrom, 2017). The Translation Challenges of Creating a New Venture in a Misaligned Institutional Context In this study we focus on Quadrant 4, which we term new venture creation through misaligned translation. Perhaps surprisingly, this type of translation has received limited attention in the literature. This is not to say that it has been ignored. For example, researchers interested in the role of meaning as a strategic resource (Rindova & Fombrun, 2001; Ravasi, Rindova, & Dalpiaz, 2012) have studied how Howard Schultz created Starbucks by translating the specialty coffee shop – an organizational form that hardly existed America – from Italy to the US when coffee retailing was in decline and the specialty coffee niche just emerging. And network sociologists have shown how scientists at US and European universities translated an organizational form – the technology firm – that was prominent in Silicon Valley to create “science-based firms” (Padgett & Powell, 2012). For example, scientists at the University of Zurich translated the concept of the technology firm to Switzerland to create Biogen, one of the first biotechnology companies (Powell & Sandholtz, 2012). While these studies have presented important insights, they have not considered the specific issues with respect to new venture legitimation. Yet theoretically we believe this type of translation is particularly interesting because the need to attain 1) local-level legitimacy and 2) category-level legitimacy raise significant problems for translated ventures, as discussed in the previous section. Attending to them pushes entrepreneurs to conform to two different sets of legitimacy pressures at an early point in their development, which can divert much needed resources from the “core product-market issues” (Fisher et al., 2016: 385) facing all fledgling ventures. The issues faced by entrepreneurs in these circumstances is complicated further when one considers the limits to conformity in the context of new venture creation: conformity is often viewed as “antithetical to entrepreneurship” (Navis & Glynn, 2011: 479) and the construction of a unique competitive positioning vis-à-vis rivals is assumed to be a prerequisite for success. Indeed, firms that mimic dominant practices precisely will be heavily penalized by stakeholders, which renders legitimacy a double-edged sword for new ventures (O’Neil & Ucbasaran, 2016). In light of the fundamental challenge posed by the need to conform and the need to differentiate, organizational researchers have increasingly viewed legitimacy through the lens of “optimal distinctiveness – positive stakeholder perception that this tension has been appropriately reconciled” (Zhao et al., 2017: 94). Although it has its roots in psychology (Brewer, 1991), a growing literature has sought to consider optimal distinctiveness in the context of organizations (Zuckerman, 2016). This work has put forward a range of possible ‘solutions’ for managing the need to be both similar and different to peers. For example, an influential idea is that organizations can seek “strategic balance” by adopting intermediate levels of deviation to become “as different as legitimately possible” (Deephouse, 1999: 147; Gardberg & Fombrun, 2006). Alternatively, there is empirical evidence which suggests that the adoption of a more radical or “pure” (Jennings, Jennings, & Greenwood, 2009) positioning involving very high levels of differentiation is effective (Zott & Amit, 2007). Other scholars have put forward more nuanced arguments for how optimal distinctiveness can be achieved. For example, firms might seek conformity to the core features of a category while differentiating on peripheral features (Wry, Lounsbury, and Jennings, 2014); they might engage in “meaning spanning” – creating and exploiting ambiguity in the minds of stakeholders (Anthony, Nelson, & Tripsas, 2017); or they might alter their approach over time by initially conforming to expectations only to deviate from them at a later point (Navis & Glynn, 2010; Zuckerman, 1999, 2016). This body of work has made important steps in addressing the issue of how firms can manage the pressure to conform and the pressure to be different. However, “there remains a great deal of ambiguity about how firms can achieve optimal distinctiveness” (Zhao et al., 2017: 94), and scholars have not looked systematically at the dynamics of optimal distinctiveness with respect to translation. These dynamics, we suggest, are especially intriguing for the specific type of translation with which we are concerned. This kind of translation requires ventures to be optimally distinct at the category-level to stand out from other, more established, members of the target form in the eyes of category-level stakeholders. However, over time, as other examples of the target form emerge in the local context, ventures may face increasing competition for resources from local-level stakeholders, and in consequence feel pressure to be optimally distinct locally. This two-fold optimal distinctiveness challenge adds an intriguing theoretical layer to our study. In sum, the creation of new ventures by translating an organizational form from a misaligned institutional context is an important and common type of translation, but one that has been largely overlooked in the literature. This kind of translation poses a fascinating set of theoretical issues, focused around the puzzle of how entrepreneurs can attain 1) local-level legitimacy, 2) category-level legitimacy, and 3) optimal distinctiveness at both local- and category-levels – a testing undertaking given that it involves managing multiple pressures from different stakeholders located in different institutional settings. To shed light on how entrepreneurs manage these pressures, our study asks: How do entrepreneurs create and legitimate a new venture by translating an existing organizational form from a different – and misaligned – institutional context? EMPIRICAL CONTEXT To address our research question, we conducted a qualitative, in-depth case study of the genesis and development of H-Farm. Single cases constitute a particularly powerful way of researching phenomena about which relatively little is known (Eisenhardt, 1989). H-Farm represents an extreme (Pettigrew, 1990) and revelatory (Yin, 1998) case because it was created through translating an organizational form (the business incubator) from a context (Silicon Valley) that was markedly different from its own (Veneto), at a time when the form itself was evolving rapidly. Below, we provide background information about H-Farm and the US incubators that were the focus of its translation efforts. H-Farm Founded in 2005 as an incubator in the digital, web and new media markets, H-Farm is located in a farming area about 30 kilometers outside Venice in the Veneto region of Italy. The founders, Riccardo Donadon and Maurizio Rossi, are local entrepreneurs who created several other ventures before starting H-Farm. However, neither Riccardo nor Maurizio had knowledge or experience of business incubation. At the time of H-Farm’s founding, private for-profit incubators were non-existent in Italy. A few appeared at the end of the 1990s as “a completely new reality in the country” (AIFI, 2001: 9), but soon went out of business during the dot-com crash (Gervasoni, 2004). Thus in 2005 H-Farm was a unique Italian venture. In the beginning, there was neither interest in H-Farm nor a vocabulary to describe it, as evidenced by scattered media coverage (8 articles between 2005 and 2006 versus 280 from 2007 to 2012), and vague labels used to describe it including “research center” (Vallin, 2007), “internet company” and “mother hen company” (Lonardi, 2005). By 2015, the number of for-profit incubators in Italy had grown to around 40 (Eit, 2015), although according to informants H-Farm remained the most high profile. By then H-Farm had evolved to combine seed incubation, venture capital investment, consulting, and education in digital technologies, and had attained considerable success. Specifically, it had secured an exit from a prominent UK-based VC, generated significant attention from the Italian political establishment and media, promoted legislative change in Italy, co-founded the European Accelerator Network, become the European advisor for the Global Accelerator Network, and was listed in the Milan Stock Exchange. Its headquarters occupied 160,000 square feet of built space and were located in two converted farm buildings. Modern, self-contained offices for the start-ups were scattered in a “seed park” surrounding the headquarters (see Figure 1). -------------- Insert Figure 1 about here -------------- The Financial Times described H-Farm as a venture that introduced to the Veneto Region, with its entrepreneurial “tradition of family businesses and family support…a new model…that you find in Silicon Valley, New York and increasingly London” (Broughton, 2012). However, several features of Veneto were unfavorable to digital incubation, such as a system of entrepreneurship that was focused on family ownership and hostile to equity investment, a lack of systematic investment in innovation, and limited use of digital technology (Marini, 2011). Crucially, Italy lacked the large and competitive VC sector that was (and is) a central characteristic of US technology regions (Crunchbase, 2017), which meant that it was a challenge for H-Farm to support the exit of its graduate firms. The Italian venture capital sector was also dwarfed by the main European innovation hubs, i.e., London, Berlin and Paris (Crunchbase, 2017). With regard to overall venture investment as a percentage of GDP, Italy had long lurked at the bottom of European rankings, a position that has not improved in recent years: in 2016 it was 0.037% in France, 0.031% in Germany, and 0.03% in the UK, while in Italy it was only 0.005%, just ahead of Romania and Greece (Invest Europe, 2017). Table 2 presents a chronology of events in relation to the creation of H-Farm. Our timeline begins in 2005, when H-Farm was founded as an incubator, and terminates in 2017, when H-Farm secured a €101 million investment for developing H-Campus – a campus for educating 3,000 students from pre-school to executive level with a focus on digital technology and entrepreneurship. --------------- Insert Table 2 about here --------------- The Evolution of Incubators in the US and Europe The first incubator appeared in the US in 1959. Since that time, this organizational form has exhibited considerable variety with respect to its goals and practices. Private for-profit incubators emerged in the US in the late 1990s. Their number rose from less than 40 in 1998 to more than 400 in 2000. By contrast, in Italy during the same period only 14 incubators had been created – all of them public and non-profit (OECD, 1999). In 2005, the year of H-Farm’s founding, a new type of for-profit incubator appeared in Cambridge, MA: Y Combinator. Its investment practices were radically different from those of existing incubators. Y Combinator’s success spawned a plethora of imitators and the attempt to “create a standardized model for their success” (http://gan.co/the-network). By 2015 the number of accelerators in the US had grown to 150 and the model had begun to evolve again (Solomon, 2015). By then, the European Accelerator Network was formed in recognition of the specific challenges for accelerating start-ups in Europe (Cruz, 2016). Further details of these developments are provided in a web appendix. METHODS Data Sources We collected interview and archival data as summarized in Table 3. Our main source of data was 53 interviews with a range of actors connected with H-Farm, Veneto, and European technology entrepreneurship. Informants included: H-Farm’s founders, partners, senior and middle managers, and startuppers; Italian policy makers and representatives of industry associations, research foundations, and other incubators; as well as Italian and European angel investors and representatives of venture capital funds. The interviews took place in three rounds (see Table 3). Overall, the interview data enabled us to build an understanding of how H-Farm was created, how members interpreted the pressures for local- and category-level legitimacy over time, and the strategies H-Farm enacted throughout the translation process. They also enabled us to triangulate why the perceived pressure to be legitimate at the category-level mattered[footnoteRef:2]. [2: We refer to interview data with a numerical code (ID#) to protect our informants’ anonymity. But we also include their role to provide context for the quotes.] ------------Insert Table 3 about here------------ In addition, we collected archival data dating back to 2005, including over 250 media articles on H-Farm, numerous publications about the Italian venture capital and incubator markets, as well as annual reports about the business community in Veneto (see Table 3). These archival data enabled us to develop an understanding of the institutional context in which H-Farm operated, the characteristics of incubation and venture capital in Italy, and how H-Farm was covered in the national press throughout its development. We also collected data about US and European incubators and accelerators, which enabled us to gain an understanding of the evolution of the organizational form that H-Farm’s founders translated, and of the features of its institutional context. In addition, we collected captures of H-Farm’s website, official videos and communication material, and the founders’ tweets during the time frame of our study, which enabled us to refine our understanding of the activities that H-Farm emphasized at different points (see Table 3). Finally, we wrote up our observations during five site visits and a meeting with representatives of the Italian investment community as ethnographic field notes. The need to keep field notes emerged during the first visit to H-Farm when one researcher, who is from the area and familiar with the business culture in the region, was struck by how informally H-Farm’s senior partners and managers interacted with her and with other organizational members and startuppers. The researcher noted that this pattern of interaction was at odds with the formal and hierarchical interactions she had experienced in other local firms. She also noted that H-Farm’s premises were rich with symbols drawn from local agricultural traditions, and that these symbols appeared central to how H-Farm perceived and presented itself. Data Analysis Our data analysis took place in several stages that involved travelling back and forth between the data, our emerging theoretical arguments, and various streams of relevant literature (Locke, 2001). For the sake of clarity, we describe the analytical stages sequentially. Figure 2 shows the final structure of our data (Gioia, Corley, & Hamilton, 2013). ------------------Insert Figure 2 about here------------------ Stage 1. Following prescriptions for case-based research (Yin, 1998), we created a database of H-Farm’s history from our interviews and archival data. Through this analysis, we established a timeline of the main changes in H-Farm’s goals and practices, and its members’ perceptions of the key challenges from its founding in 2005 to 2017. We integrated the timeline with data about the regional and national context in which H-Farm was created, as well as with data about the evolution of the incubator model in the US and Europe. This narrative allowed us to develop a detailed understanding of a) the evolution of H-Farm’s goals and practices; b) organizational members’ interpretation of the meanings attached to such goals and practices; c) organizational members’ perceptions of the different kinds of pressures faced by the venture and of their intensity, and d) the features of the institutional contexts in which H-Farm operated. Stage 2. Next, we engaged in a first-order analysis (Van Maanen, 1979) that involved a thorough open coding of the interviews and field notes. Following prior research (e.g. Corley & Gioia, 2004), we used sentences or paragraphs as coding units and labelled them with either in vivo codes or descriptive sentences. Using the constant comparative method, we repeatedly compared data over time and across informants to discern the major concepts of interest. The first-order concepts helped unveil informants’ understandings of 1) the actions that contributed to H-Farm’s genesis and development, 2) the legitimacy dynamics that H-Farm encountered, and 3) the changes to the target form over time. For those first-order concepts focused on actions, we came to the view that some were concerned with choosing and refining specific practices that allowed H-Farm to function, i.e., “practice work” (Gawer & Phillips, 2013), while others were concerned with constructing and ascribing meaning to H-Farm’s activities; i.e., “meaning work” (Rao & Giorgi, 2006). Stage 3. We observed that the founders’ understanding of incubators changed over time. Consistent with historical case analysis (e.g., Nasra & Dacin, 2010), we analyzed these changes systematically by looking for “critical junctures” in H-Farm’s development, defined as points in time that “durably transform previous structures and practices” (Sewell, 1996: 843). We identified three such junctures and provisionally bracketed (Langley, 1999) the timeframe of the study into three discrete periods – during each of them we observed that the founders used the concept of an incubator as a model for their venture, but in different ways and referencing different incubators, and that they perceived the legitimacy challenges differently. We then used these periods as embedded units of analysis (Eisenhardt, 1989) to theorize the process of creating and legitimating a new venture through translating an organizational form from a misaligned context. Stage 4. We next aimed to uncover deeper patterns in the data. Through axial-coding we established links among the first-order concepts in each period to derive second-order themes (Locke, 2001) that described the legitimacy pressures experienced, how the legitimacy pressures were tackled, and the entrepreneurs’ changing understanding of the target form. To do so we again used constant comparison techniques. The second-order themes represented our working hypotheses that we tested against available data (Glaser & Strauss, 1967). Seven second-order themes emerged from this stage of analysis. Stage 5. We then collapsed the second-order themes into three aggregate dimensions that captured the overarching constructs that represent the theoretical building blocks of the type of translation with which we are concerned (see Figure 2). Stage 6. In the last stage of our analysis, we drew multiple representations of the emerging theoretical relationships among constructs and themes within and across periods, and recursively checked with the data to ensure consistency (Locke, 2001). We then developed a theoretical model designed to illustrate diagrammatically these relationships over time. THEORETICAL MODEL: OVERVIEW Our model of new venture creation through misaligned translation, shown in Figure 3, discloses how entrepreneurs create a new venture by translating an existing organizational form from an institutional context very different to their own, and the changing legitimacy pressures encountered along the way. The constructs upon which the model is based emerged inductively from our grounded theory building. However, we briefly introduce the model here to preview and help structure our findings. The model begins with a first translation phase, which we label improvising, in which entrepreneurs locate an organizational form – an initial target – in a misaligned institutional context that they aim to loosely emulate. In seeking to transfer that form to their own context, they immediately face questions from local-level stakeholders about the legitimacy of their new venture. To respond to these perceived local-level legitimacy pressures, entrepreneurs engage in what we term local-level authentication work – they seek to explain the venture to local-level stakeholders and adjust the target model so that it fits with local expectations in an effort to secure access to local resources. Perceived changes to the target organizational form trigger reflection (Dalpiaz, Rindova & Ravasi, 2016; O’Neil & Ucbasaran, 2016) on the part of entrepreneurs. Where, as in our case, entrepreneurs believe there is a need to acquire resources outside of their locale, a second translation phase, which we label converging, is precipitated in which entrepreneurs seek approval from stakeholders of the target form. In doing so, entrepreneurs face questions from these category-level stakeholders, which are located in another institutional context, about the legitimacy of their venture and whether it is a ‘real’ category member. To respond to these perceived category-level legitimacy pressures, entrepreneurs engage in what we term category-level authentication work – they seek to explain the venture to category-level stakeholders and conform to their expectations in an effort to secure access to resources from them. At the same time, they continue to perceive pressure to conform to local-level legitimacy pressures, and to engage in local-level authentication work, but in this phase the two sets of processes occur in parallel. Further perceived changes to the target form again trigger reflection on the part of the entrepreneurs. Where, as in our case, the limits of convergence towards a ‘standard’ category become apparent to entrepreneurs, a third translation phase is precipitated, which we term optimizing. At this point in the translation process, our model suggests that the perceived local- and category-level legitimacy pressures evolve. At the local-level, the target form may now be accepted and widespread, which means that as well as conforming to the expectations of local-level stakeholders, the venture needs to differentiate itself from local rivals. At the category-level, the target form may be widely adopted in many institutional contexts, which means that as well as conforming to the expectations of category-level stakeholders, the venture needs to differentiate itself from international rivals. To respond to these twin sets of pressures and to become accepted but different at both the local- and category-levels, our model suggests that entrepreneurs engage in what we term dual optimal distinctiveness work. This entails a kind of inversion vis-à-vis the previous phases: the actions underpinning local-level authentication work are used to legitimate the venture to category-level stakeholders, while the actions underpinning category-level authentication work are used to legitimate the venture to local-level stakeholders. ------------------Insert Figure 3 about here------------------ The next section details our case analysis in which we ground our theoretical constructs in our data, and explain the relationship between them. NEW VENTURE CREATION THROUGH MISALIGNED TRANSLATION: THE CASE OF H-FARM Phase 1 (2005-2009): Improvising H-Farm was created in 2005 as a venture to support technology entrepreneurship in Veneto – “a platform to help young people to start initiatives on the internet” (ID14). The founders, Maurizio and Riccardo, were captivated by technology entrepreneurship in Silicon Valley. They knew that a type of organization – the incubator – was prominent there, but lacked detailed knowledge or direct experience of business incubation. In other words, they had only a generalized idea about what incubation in the US entailed rather than a specific model in mind. Indeed, there was no ‘standard’ model of incubation for H-Farm to emulate: the non-profit and for-profit incubators in the US, which formed the reference point for H-Farm during this first phase, were characterized by a diversity of practices (see web appendix). One of the founders explained that there had been little forward planning: “when we started… we did not undertake any analysis: just the intuition that there would [be] a market [for digital businesses in Italy]…we started off, and adjusted our path along the way” (ID27). We therefore labelled the first translation phase improvising. Crucially, no private incubators existed in Italy at this point, and the institutional context of entrepreneurship in Veneto seemed both detrimental to the new venture’s growth, and at odds with some of the fundamental characteristics of digital entrepreneurship. More fundamentally, local-level stakeholders – entrepreneurs, investors, and policy makers – were not familiar with the concept of a private incubator, and were not therefore willing to provide it with resources. This meant that H-Farm’s founders had to convince these stakeholders that their venture was worthy of support. In this section, we first describe the local-level legitimacy pressures faced by H-Farm. We then describe the translation legitimacy work – which we label local-level authentication – in which H-Farm engaged to address these challenges. Note that our analysis suggests that the perceived pressures for category-level legitimacy (i.e., legitimacy among stakeholders of the target form) are minimal during this phase – the primary objective is for the venture to become accepted locally so that it can take root; conforming to the category is less of a priority in this respect because the form being translated is unfamiliar in the local context. Perceived Pressures for Local-level Legitimacy Our analysis suggests that, initially, the core challenge faced by entrepreneurs creating a venture through misaligned translation is attaining local legitimacy. This is because local-level stakeholders have no experience of the venture and therefore struggle to relate to it. It is a challenge that becomes particularly acute when, as in our case, the local institutional context is very different from the context where the target form emerged – entrepreneurs need to explain the nature and purpose of the venture so that it ‘makes sense’ locally. But in doing so, they face pressures to position the venture in terms of existing understanding about entrepreneurship. In the case of H-Farm, there were three main local-level stakeholders from which the venture sought legitimacy. The first was local entrepreneurs and prospective entrepreneurs. Unlike in established technology regions in the US and the UK, there was no culture of technology entrepreneurship in Veneto or Italy. Indeed, digital business was in its infancy – it was widely perceived as disconnected from, and much less important than, the traditional forms of economic activity in the region where small and medium manufacturing firms dominated the local economy (ID21; ID22). As a result, H-Farm’s founders said that few young people considered it as a viable way to make a living, preferring instead the stability of more traditional career paths. This, the founders said, affected profoundly the amount and quality of potential startuppers: “Italy is not like... London or Cambridge [where you have] young, tech sector guys... tech programmers looking for jobs” (ID 16). This view was supported by the OECD (2014), which argued that “an ageing of entrepreneurs in Italy” has long represented a key innovation challenge in the country, and that “orienting youth… towards the possibility of entrepreneurship” should be a key priority (p. 159). Without legitimacy in the eyes of local entrepreneurs and prospective entrepreneurs, H-Farm’s founders reasoned that they simply could not build the deal flow required to be a viable incubator. The second key set of local-level stakeholders from whom H-Farm sought legitimacy was local investors. The venture capital sector was (and remains) very under-developed in Italy with banks, and to a lesser extent established firms and private investors, the main providers of finance for entrepreneurs. The founders emphasized that when H-Farm was created Italy lacked the financial intermediaries that could invest in H-Farm’s fund and graduate start-ups. This perception was corroborated by other data we collected. For example, economists noted that in H-Farm’s formative years “Italy lacked the segment of early stage financing and has been characterized by the lack of a culture for early stage [investment]” (Gervasoni & Sattin, 2008: 107). External experts confirmed that the limited VC market represented a major problem for start-ups in Italy, including H-Farm (ID38). Il Sole 24 Ore, the Italian business daily, reckoned that in the absence of a developed financial infrastructure, the founders’ “real gamble is pursuing this innovation path in Italy rather than Palo Alto…[as] the comparison with the US is ruthless” (Marrofoto, 2009). Without legitimacy in the eyes if the Italian investment community, particularly the banks, large firms, and wealthy private investors, H-Farm’s founders did not believe their venture was viable. The third key group of local-level stakeholders from whom H-Farm’s founders sought legitimacy was policy makers. As noted, the institutional context of Veneto and Italy was not conducive to the emergence of H-Farm, which represented “an effort to create a technology and venture capital hub in a country with neither” (Broughton, 2012) and in a region that “lack[ed] an eco-system that could be favourable to innovation” (Seganfreddo, 2012). Several features of the Italian entrepreneurial environment hindered the creation and growth of a digital incubator, including a legal framework that was focused on family ownership and hostile to equity investment, a lack of systematic investment in innovation, limited investment in, and use of, digital technology, as well as the issues around finance discussed above (Marini, 2011). H-Farm’s founders were exasperated at policy makers’ apparent apathy towards technology entrepreneurship, and their concerns were supported by the OECD (2014: 160) who noted the lack of a “national strategy and action plan” in this regard stretching back over many years. Thus, at a very early stage, H-Farm’s founders believed that they needed to build their venture’s legitimacy in the eyes of government. Local-level Authentication Work In order to overcome the local-level legitimacy challenges in the translation process as outlined above, we found that entrepreneurs engage in local-level authentication work which, building on Howard-Grenville et al. (2013), we define as the process of constructing resonance between the new venture and its local context so that the new venture is perceived by local audiences as authentic, and hence accepted. The aim is to convince local-level stakeholders that the venture is “appropriate” and “desirable” (Suchman, 1995) in the local context. We found that local-level authentication work comprises 1) practice work focused on the concrete construction of the new venture in its local context – the ‘nuts and bolts’ of how it actually works; and 2) meaning work focused on creating a strong association between the new venture and elements of an established local meaning system. Practice Work. Riccardo and Maurizio’s assessment of the profound differences between the Veneto region and established US technology regions led H-Farm “to try to find out the good things [about Veneto]” (ID13) and to develop practices appropriate for that context rather than to focus on what incubators in Silicon Valley did, which in any case was fragmented at this point. Moreover, as Veneto appeared so unfavourable to technology entrepreneurship, they reasoned that any attempts to replicate precisely the goals and practices of US incubators would not succeed. Consequently, the founders said they needed to focus on practices that fitted their context regardless of what ‘real’ incubators actually did. As a result, during its early years H-Farm improvised practices that had little to do with those of its American counterparts (see Table 4). For example, H-Farm initially owned equity stakes of 90% of the incubated start-ups. This differed from the US where incubators usually took a minority stake, or no stake, and had a significant impact on the relationship between mentors and startuppers. In addition, the tenure of the ventures in H-Farm’s portfolio was intended to span several years rather than the shorter periods common in US incubators; there was no clearly proscribed incubation period. And, unusually, the identities of the start-ups were closely connected to H-Farm (the names of the first 3 ventures were even preceded by “H-”). Indeed, at this point H-Farm could more accurately be described as a “holding company” of majority interests (ID27) than an incubator of start-ups. For example, a partner noted that: “H-Farm as main investor meant that H-Farm, in fact, was the manager of the start-up, which led to a relationship of ‘headquarter with subsidiaries’ rather than ‘accelerators with many projects that could stand on their own feet’” (ID25). Some of H-Farm’s practices did more closely resemble US incubators. For example, recognizing the shortage of entrepreneurs in Veneto, it created networking events designed to build interest and ultimately deal flow, as happened in Silicon Valley. For example, it introduced the practice of a weekly dinner for startuppers and others from the local business community, labelled “Storming Pizza” – pizza being associated with informal dining in Italy. The aim was to allow prospective entrepreneurs to network and compare ideas, and to meet people who had already created a venture. In general, however, because H-Farm started out translating the broad idea of a US incubator rather than a specific set of practices, at this stage it was “an incubator…in quite a light way” (ID14). Meaning Work. As well as differing significantly from US incubators, H-Farm was also highly idiosyncratic in Italy. Consequently, the founders believed they needed to create resonance between H-Farm and its local context so that it was accepted as legitimate in the eyes of local entrepreneurs, investors and policy makers. Thus, as well as assembling a distinct set of incubation practices designed for Veneto, the founders engaged in meaning work that was intended to make these practices – and the organization more broadly – resonate locally. They did so by linking H-Farm with the history, symbols, language, and materials that characterize Veneto and Italy. First, Riccardo and Maurizio sought to construct a relationship between H-Farm’s activities and Italian heritage by positioning the venture as the embodiment of a “humanistic approach” to innovation (the “H” in H-Farm). Humanism emerged in Italy during the Renaissance (around the 14th Century). It is a cultural and educational perspective that prioritizes creativity, ingenuity and non-conformity. One of the founders described humanism as “our philosophy, the idea [underlying] the Human Farm” (ID13). He said that a focus on humanism represented an opportunity for H-Farm because it orients people towards a specific type of innovation – “human-centric innovation” – that leads to simplified and aesthetically appealing products and services that are designed around the users’ experience. Second, H-Farm drew on regional and national symbols to underline its human-centric approach. For example, signposts in the park, brochures and publications, logos, and the firm’s website featured farm-related symbols: tractors, farmers, and hoes. Informants stressed that this was not a “marketing manifesto” (ID10). Rather, it underlined the role of humanism in the organization’s approach to innovation: This [farmland] used to be for farming and now it’s for digital farming. We’re growing digital products, so it’s more consistent being here because we’ve been in a place where there was the farm, and the farm had also the sense of creating a house for entrepreneurs… We have metaphors, double senses, but everything stays really driven by the strategy [of] making the technology around the human, and human-inspired [technology] creates something different (ID10). Third, H-Farm attempted to create resonance with the Veneto region by highlighting stories of successful local entrepreneurs, both contemporary and historical. These stories, combined with opportunities to meet the entrepreneurs in person, were intended to inspire startuppers and imbue them with self-belief about what they could achieve. The focus on local rather than international examples was deliberate: H-Farm’s founders believed such entrepreneurs were much more likely to be perceived as imitable. An informant elucidated this point evocatively as follows: Learn from Marco Polo…not Zuckerberg! …The first publisher to use copper printing plates in the world… was Zatta from Venice. … So that’s the point. We need to find something in our circle…These are examples from our culture and we're not so ready for…big investment, big funding. We can try to leverage on our [local] opportunities (ID13). Finally, H-Farm attempted to use material features to create resonance between the venture and the local context. The aesthetic qualities of H-Farm’s environment – beautifully designed farm buildings in the scenic Italian countryside – were linked to Veneto’s artistic tradition and its close association with humanism. The following excerpt illustrates how Riccardo sought to frame these material features of H-Farm so that they resonated with material features from the local context: The humanism in Veneto, which…expressed masters such as Bellini and Mantegna, is a way to use the gaze to understand things. It is a sense of the landscape and of architectures, which I carry as a heritage [from my studies in the humanities]….H-Farm is built on ample, open and airy spaces. It seeks a deep balance with the surrounding micro-cosmos. Indeed, with H-farm I had three objectives: be respectful of the landscape, create a beautiful thing, and render concrete the digital thinking (in Bellini, Donadon, Costa, 2013: 31-33). The founders became very skilled at communicating these messages to local-level stakeholders. They did this partly through local events targeted at entrepreneurs, as well as through social media, and were helped in this respect because they were very well connected: Riccardo had contacts in the highest echelons of government, and both founders had strong relationships with the Italian business and investment community. Transition from Phase 1 to Phase 2 In summary, during phase 1 we observed that the founders created their venture by drawing on the US incubator model in a very loose sense, in part because there was no coherent form to emulate and in part because were significant differences between the context from where the form originated (Silicon Valley) and the context in which it was being translated (Veneto). As a result, the entrepreneurs improvised with alternative practices designed to ‘fit’ the local context. The desire for local legitimacy was prioritized over fidelity to the original form, which was in any case ill-defined in their eyes. Perceived Target Shift Towards the end of Phase 1, H-Farm’s founders believed that the target form had shifted. Moreover, although still characterized by tremendous diversity, US incubators were viewed by H-Farm as evolving in a way that was more transparent (see web appendix). Most notably, Y Combinator and Techstars emerged as high profile exemplars, attracting significant media attention. Riccardo and Maurizio read a blog about Y Combinator – founded in the “same year and month that H-Farm was born” (ID27) – which was pioneering a new incubation model. It caught the attention of H-Farm’s founders. According to one of the founders: This was the first time we heard the word ‘accelerator’. We thought it was a new generation of incubator, with specific rules that are [the following]: positioning at the very early stage of the start-up’s life cycle; giving three months to prototype the business model; offering the output to external investors that, if they are interested, invest in the venture to make it take off. This thing actually intrigued us (ID27). Crucially, in 2010 Techstars founded the Global Accelerator Network (GAN), which was designed to disseminate a set of “best practices” for accelerating start-ups (see web appendix). H-Farm’s founders began monitoring the activities of the accelerators that belonged to the GAN by reading specialized articles in blogs and websites. Riccardo explained the purpose of doing so as follows: “Certainly, sometimes having a look at how others are organized helps you to keep the trajectory and being comforted about your own model…to understand whether it is right or not” (ID28). This rendered the original form less opaque, and made the task of learning about it easier than hitherto. Thus, the emergence of Y Combinator and Techstars and the creation of the GAN precipitated a change in H-Farm’s focus from a generalized model of incubation, to a more specific model based around an identifiable set of practices as promoted by this new generation of American incubators, and which became the primary reference point for H-Farm’s translation efforts. Reflection The perceived changes to the target form led H-Farm to reflect deeply on its own model. By the end of 2009, it had supported 9 ventures and successfully exited the first three start-ups it invested in, all of which were sold to industrial groups and generated an internal rate of return between 140 and 165%. These positive outcomes corroborated the founders’ belief in H-Farm’s potential. Yet, they also became aware of several critical management challenges that revealed the limits of its way of operating. First, the practice of taking majority investments in its start-ups meant that H-Farm’s capital was tied up in a narrow range of ventures. Second, exiting proved challenging as the traditional options exercised by US incubators – IPOs, selling to VCs, and merger and acquisitions deals – “did not exist” (ID27) in Italy. Third, raising capital from external investors was difficult because incubation activities were not well understood in Italy. H-Farm was forced to rely on “friends and family” (ID13) for capital, but the relatively small amount of capital that each conferred (totalling €3 million) constrained H-Farm’s investment in start-ups. H-Farm’s founders felt that the challenges they encountered were rooted partly in their location in Veneto. Put simply, there was neither the deal flow in terms of entrepreneurs nor the investors to support lucrative exits. They came to the view that to be successful H-Farm needed to attract talented entrepreneurs from around the world, and to persuade international VC firms to invest in the ventures it was incubating. Riccardo and Maurizio therefore began “reflecting on a new architecture for H-Farm” (ID27) that made it “a very different venture” (ID01) than during the first phase, and moved it closer the US model of incubation. Phase 2 (2010-14): Converging The belief that H-Farm could not access the requisite entrepreneurs and investment from Veneto, combined with the founders’ increasing awareness of the apparent limitations of H-Farm’s own emerging model, triggered a new translation phase, which we label converging. In this phase, the central pressure involved attaining legitimacy in the eyes of category-level stakeholders; i.e., acceptance amongst international entrepreneurs, investors, and incubators. To address these challenges H-Farm felt that it had to become more like the leading US incubators, which were held up as exemplars. They therefore evaluated the original form differently than before – whereas they had initially experimented in a rather casual and haphazard way, they now began to more carefully select particular elements of the target form, focusing on Y Combinator, Techstars, and the GAN, and to graft them on to their venture intact. Although, local-level legitimacy challenges remained and local-level authentication work continued during this second phase, the focus of attention switched to making H-Farm understood at the category-level. Perceived Pressures for Category-level Legitimacy Our analysis suggests that as a new venture that has been translated from a different institutional context matures, a new translation challenge may emerge, namely, the need to attain category-level legitimacy – legitimacy among key stakeholders associated with the target form. This is because these stakeholders may hold access to resources required for growth that are absent in the local context. Again, it is a challenge that is particularly acute when, as in our case, the local context is very different from the context where the target form emerged. The importance of international recognition for H-Farm was highlighted in several of our interviews. For example, an informant at a high-profile VC firm in London told us the following: Is international recognition important [for Italian incubators]? Yes, if you want to have exits. You will have better deal flow, you will have better people involved… If you expand internationally, you will have better people coming in the door, and better people in house giving harsh, honest feedback and opening doors. The better link with international investors, the better chances of getting exits (ID39). In the case of H-Farm, there were three main category-level stakeholders from whom H-Farm sought legitimacy. This first was international entrepreneurs. There was widespread acceptance that Italy suffered from a shortage of good quality entrepreneurs, which severely limited deal flow. This perception was confirmed by the same London-based informant: “If you are a European start-up and have a choice where to go, you would not go to H-Farm.... Being in Italy is a disadvantage from day one” (ID39). The problem was exacerbated by the number of young Italians leaving the country to create their ventures elsewhere, especially London and California. For example, the co-founder of Hyperfair – a highly successful digital business – reluctantly chose to set up his venture in Silicon Valley rather than Italy because he felt it had a much better chance of success there: “I am Italian and I love Italy. But… Back in Italy, each day began with a list of the problems I had to solve. Here I make each day a list of the things I want to do” (CNBC, 2013). H-Farm believed that it needed to attract entrepreneurs from other countries to make up for the lack of local talent. However, here they were at a significant disadvantage: there was a widespread perception that Italy was not a ‘natural’ place for digital entrepreneurship. The second group of stakeholders from which H-Farm sought legitimacy was international investors. As noted, the venture capital industry in Italy is very under-developed. H-Farm had been working with high-net worth individuals, but had fallen well short of the investment that the founders believed they needed. They therefore took the view that they had to attract international VC, particularly from London and Berlin, the two main centers of European venture capital. However, as with their attempts to lure international entrepreneurs, their location in Italy appeared to put them at a significant disadvantage for attracting such investment – the founders said they became frustrated that H-Farm was often discounted by investors who had only limited knowledge and understanding of it. As one international investor commented, “we do not look at Italy as a source of start-ups…Italy is not on [our] radar. It lacks government support...that could act as a megaphone for the quality of Italian startuppers….It lacks a national identity [in technology]. It should develop an identity in fashion tech or food tech for example...Then, [Italy] would earn credibility and [international] investors would consider meeting [Italian] entrepreneurs” (ID40). Finally, H-Farm sought legitimacy from the most high profile accelerators in the world, particularly those in the US. While H-Farm did not expect these peer incubators to provide resources directly, they believed that recognition from them would help build its reputation as a ‘serious’ accelerator, which would facilitate access to international entrepreneurs and investors. But gaining acceptance among this group was challenging: H-Farm was largely unknown outside Italy, as evidenced by the fact it had only 6 mentions in the English speaking media up to this point, all with reference to one of its incubated start-ups with headquarters in Seattle. Again, H-Farm’s location in Italy appeared to work against them. To the extent that the high profile Silicon Valley incubators sought to engage with European partners, they looked mainly to London and a small number of other European cities. According to a senior manager at H-Farm: “there are three tiers of entrepreneurial ecosystems in Europe. The ‘A level’ ecosystems are Berlin and London. The ‘B level’ ecosystems are Paris, Stockholm and Helsinki. The ‘C level’ ecosystems are Italy and Spain” (ID48). This perception was reinforced by the behavior of leading incubators. For example, Techstars chose to focus its European start-up programs in London, Berlin and Paris. Similarly, Seedcamp, the largest European accelerator program that draws on investors from across Europe, chose to locate in London and Berlin. Category-level Authentication Work To overcome the category-level legitimacy challenges in the translation process, outlined above, we found that entrepreneurs engage in category-level authentication work. We define category-level authentication work as the process of constructing resonance between the new venture and the target organizational form on which it is based so that the new venture is perceived as authentic, and hence accepted, amongst stakeholders associated with the target form. The aim is to convince category-level stakeholders that the venture is an “appropriate” and “desirable” (Suchman, 1995) member of the target organizational form. We found that category-level authentication work comprises 1) practice work focused on the development of practices that conform to those of the original form; and 2) meaning work focused on creating a strong association between the target form and the new venture. Practice work. As a first step in their efforts to attain category-level legitimacy for their venture, H-Farm’s founders decided to compare H-Farm much more systematically with established accelerators – to immerse themselves and become expert in the way that US incubators were organized – to help build a clearer vision for how the translated venture should develop. In other words, they sought to “scout models” and “check that we have taken a compatible trajectory with the evolution of the market [i.e., other accelerators]” (ID28). The benchmarking was facilitated by the increasing quantity of data being published by the GAN. A partner who had prior experience in VC joined H-Farm to help with the analysis. He commented that they “looked in general at a lot of different models and…at what we thought would work in Italy” (ID16). The goal of this analysis was “to create a method that is less naïve than [the one] just based on inspiration, [a method which is] more pragmatic instead, more…a method indeed!” (ID10). As a result of this exercise, a new set of practices were implemented that more closely aligned H-Farm to their evolving perception of the target form as exemplified most obviously by the GAN (see Table 4). For example, H-Farm abandoned majority investments (70-90%) and moved to minority investments (25-30%). A partner defined this shift as “a turning point towards a way of working that is more like accelerators” (ID25). Moreover, in 2011 H-Farm launched the first Seed Program. This included the practice of setting a deadline for collecting proposals, which is standard among the GAN members (see web appendix). In 2012, the Seed Program evolved again: it featured even lower stakes (from 25-30% to 10%) and cash investment (from 30k to 15k), as well as “batch investments”. H-Farm members described these features as recreating “in a more precise way” (ID31) the typical practices of US incubators. In addition, H-Farm moved from a “fund-like structure” (ID26) where each partner took responsibility for one start-up, “followed it very closely” (ID25) and sat on the board, to a “mentor-like structure” composed of a team providing day-to-day support, and a network of 30-40 mentors that startuppers could turn to for advice on an ad hoc basis. Again, this change was described as a “turn towards a structure of accelerators” (ID26) and the GAN was explicitly referenced as the inspiration for these changes. ----Insert Table 4 about here----- H-Farm also sought to cultivate the norms of collaboration commonly found in US incubators. Specifically, they sought to create a supportive environment inside the venture, in part through co-location, to promote the sharing and cross-fertilization of ideas. Crucially, sale agreements included a provision for the start-ups to remain at H-Farm to help develop a critical mass of digital experts. They promoted the practice of working in the ‘Greenhouse’ – an architecturally impressive meeting and eating space so named because of its aesthetics (glass and steel structure and plants inside) and because of its intended function, which was to cultivate the development of ideas. According to one of the founders: Our Greenhouse…corresponds to a Starbucks in America, where you can go with your computer... I… tried to recreate an internal Starbucks where we all go, have coffee, and I talk to the people who are working in the start-ups. I tried to recreate something like those American malls that recreate an old…village (ID14). But while there was clear convergence, there remained differences between the practices enacted by H-Farm and its US counterparts: H-Farm was unable to replicate the US incubator model precisely in Veneto. For example, H-Farm engaged in both seed and follow-on investments. The need to extend incubation to the post-seed stage was due to the lack of third-party investors with the resources to invest in incubated firms. According to one of the founders: We tried to understand how an accelerator could work in Italy, given the lack of external investors – because an accelerator makes sense if it can invite 100 investors at the “demo day”…with 100 investors you are able to exploit the investment you made, otherwise regardless of the interest of the initiatives you accelerated, if you cannot present them to anyone, you either let them die or take responsibility to invest further in some of them. So, we created an investment company that acts as a third-party investor (ID27). This practice was not, our informants told us, found in US incubators. As a partner remarked: “we do not do like Y Combinator and Techstars that do the first step and then they stop…H-Farm does follow on some start-ups” (ID25). Meaning Work. In addition to creating practices that conformed to the US model, H-Farm began to use language that mimicked that of US incubators. For example, H-Farm split into “the Accelerator” that focused on seed investments in start-ups with minority stakes, and “H-Farm Ventures” that focused on follow-on investments. Thus, the new structure assumed the labels and goals of acceleration and VC common in Silicon Valley. The founders also embraced the distinctive communication style of Paul Graham, Y Combinator’s founder, because they recognized that startuppers, who are usually young programmers, require “a very direct, informal and pragmatic language” (ID28). Second, H-Farm sought to affiliate with the community of US and other international accelerators. This was particularly noticeable from 2012, when H-Farm applied to be admitted to the GAN. Indeed, H-Farm co-organized with Techstars the first GAN Meeting, which was hosted at H-Farm – a crucial opportunity to stake its claim as a ‘real’ member of the incubator category. In early 2013, H-Farm was deemed to meet the requirements for GAN membership, and formally joined. The second GAN meeting was hosted at H-Farm in the spring of that year. The founders saw GAN membership as important not only for learning about “best practices”, but also for building the legitimacy of H-Farm among its peers. Finally, while clearly a longer-term project, the founders wanted to develop the reputation of Veneto as a hub for digital entrepreneurship. While this was motivated partly to improve the performance of H-Farm, fundamentally it represented an attempt to put H-Farm on the map – a way of getting other international stakeholders to take notice of what was happening in entrepreneurship there. As a partner explained: “the idea is to create an eco-system” – not a Silicon Valley, but a “Sile Valley” (ID05) – the Sile is the name of the river bordering H-Farm. Transition from Phase 2 to Phase 3 To summarize, in Phase 2 a new dimension to H-Farm’s translation efforts emerged: perceived changes to the target form led H-Farm to reflect on its own model and to look beyond the country’s borders for resources to compensate for the limited number of startuppers and restricted supply of capital in Italy. In doing so, H-Farm began to seek legitimacy from, and conform to, key category-level stakeholders – international entrepreneurs, investors and incubators. This did not mean that H-Farm’s efforts to be legitimate locally ceased – on the contrary, the founders were cognizant that they could not rely solely on international resource providers. However, the two sets of legitimation processes – local-level and category-level – largely occurred independently of one another; i.e., they were compartmentalized. Perceived Target Shift Towards the end of Phase 2, H-Farm’s leadership team again perceived that the target form had shifted. Specifically, the scale at which US incubators were operating, and the level of investment in the start-ups appeared to have increased markedly. For example, typical investments at Y Combinator were around $120,000, and a range of additional – and costly – support services were offered to start-ups. At the same time, the number of European accelerators grew rapidly. While those in the UK sought to emulate their US counterparts, in continental Europe accelerators were characterized by a diversity of practices. In 2013, the European Union funded ATALANTA, a project to promote European entrepreneurship, with “a special focus on acceleration programs, the new paradigm regarding entrepreneurship support” (NUMA, 2014). Seven European incubators – H-Farm and six others – were members of ATALANTA. The project came to an end in 2016, but had a profound impact on European incubation. Specifically, “a significant outcome” of the project was the creation of a new network of accelerators – the European Acceleration Network (EAN) launched in 2014. In other words, a European version of the GAN. H-Farm was a founding partner, but the network grew quickly to 24 members. Its core objective was as follows: “This network aims to strengthen acceleration in Europe by bringing accelerators together in an open, collaboration network to exchange best practices and explore opportunities. This networking opportunity will benefit directly the accelerators who will learn and implement new methodologies, establish new connections but will also be a catalyst for growth of acceleration in Europe” (Atalanta, 2016). Crucially, the creation of the EAN precipitated a shift in H-Farm’s focus: the founders’ primary reference point changed from Techstars, Y Combinator, and the GAN to the EAN. This altered their perspective on the incubator model yet again, and focused their attention upon a broader set of practices for accelerating start-ups, as promoted by the EAN. This did not mean that H-Farm ignored what was happening in the US – they continued to be active in the GAN and to follow the US scene closely – but the EAN was deemed more relevant to the specific challenges and opportunities of being an incubator in Europe. Reflection The emergence of the EAN led H-Farm to reflect once again on its progress and model. H-Farm’s leadership believed much had been achieved. By the end of 2013 it had incubated 67 start-ups, one of which had attained investment from a high profile London-based VC firm. In addition, it had secured €10 million of additional funds, to be invested between 2013 and 2018. And there were now 353 people based on at the H-Farm campus (startuppers and staff) spread across 86,000 square feet of land. And yet significant problems remained. Despite their growing international profile, H-Farm was not attracting many entrepreneurs from outside Italy, nor were they attracting the level of international investment they had anticipated – the hoped-for influx of global talent and capital did not materialize. H-Farm did believe that significant progress had been made locally, but, as noted below, a new wave of Italian incubators had begun to emerge which meant increased competition for local entrepreneurs and capital. At the same time, there was growing scepticism among EAN members about the applicability of the US incubator model in the European context, at least outside of London and Berlin. A British investor noted bluntly that “there is a problem with the whole model” (ID41) of US acceleration and incubation. This led H-Farm to conclude – which we interpreted as something of an epiphany – that the US model simply did not work in Italy. As a result, H-Farm believed that it could not continue to move ever closer to the US incubator model if it was to thrive in the long-term. Phase 3 (2015-17): Optimizing The belief that convergence towards a US incubator model was unsustainable triggered a third translation phase, which we label optimizing. In this phase, H-Farm continued to face the two main translation challenges of building legitimacy among local- and category-level stakeholders. However, it perceived that the nature of these challenges had changed because of the rapid expansion of incubators both within Italy and across the world. This led to a significantly more crowded incubator landscape and a marked increase in competition for resources. H-Farm therefore took the view that, not only did it need to gain acceptance from local- and category-level stakeholders, it also needed to differentiate itself from other incubators in the eyes of these stakeholders. In other words, H-Farm needed to become optimally distinct at the local- and category-levels. Perceived Pressures for Local-level Legitimacy Our analysis suggests that the local-level legitimacy pressures changed markedly during Phase 3. In a key respect, these pressures had lessened because the incubator form had become more accepted in Italy. Indeed, the number of private incubators in Italy grew rapidly – from just a handful at the beginning of Phase 2 to 40 by the beginning of Phase 3. As the first private incubator in Italy, H-Farm believed that it had played an important role in this respect, but the generally shifting local landscape and the attention given to US incubators such as Y Combinator in the Italian media also appeared to have played an important role (Atkinson, 2016). Crucially, government legislation, passed in 2012, explicitly recognized the importance of incubators for the Italian economy. Riccardo, one of H-Farm’s founders, was part of the “Restart Italy!” Task Force, which helped the Ministry for Economic Development draft the bill. This led to a number of changes to Italian law designed to support entrepreneurship in general, and incubation in particular, including a visa program to attract international entrepreneurs, legislation that allowed for stock options, and the easing of tax and other regulations for incubated start-ups. In addition, AIFI – the Italian industry association for venture capital – created an initiative called “VentureUp” to promote technology entrepreneurship. In light of these changes, H-Farm’s leadership believed that local-level stakeholders – entrepreneurs, investors and policy makers – all had an increased understanding of incubation (i.e., the incubator as a category was increasingly legitimate at the local-level). This meant that H-Farm moved from a situation where, as the first and only private incubator in Italy, it was seen as “something bizarre, almost folkloristic” (Bellini et al., 2013: 29) to a situation where there was broad recognition of what an incubator is for and support for its existence. H-Farm therefore felt less need to explain its goals or defend its approach. The growth in the number of start-ups that H-Farm had incubated and the marked increase in the investment it had attracted, outlined above, provided evidence of its growing local-level legitimacy. However, the explosion in the number of private Italian incubators created a new local-level legitimacy problem. Specifically, H-Farm now had to compete with these incubators for resources in the form of entrepreneurial talent as well as capital from investors. This meant that H-Farm had to develop a positioning strategy that situated itself as part of this new and exciting wave of incubators in Italy, but at the same time was different from them so that it could ‘win’ the competition for local resources. This was crucial because, despite its efforts to attract resources from category-level stakeholders, H-Farm continued to rely heavily on entrepreneurs and investment from inside Italy. In other words, H-Farm needed to become optimally distinct locally. A senior manager summarized the situation as follows: “When H-Farm started we were the first, there were not a lot of accelerators slash incubators in Italy. And it’s obviously true that there has been a huge expansion on that. So H-Farm… [has] always asked: how can we be different?” (ID49). It might be expected that the growth in the number of Italian incubators simply reflected the growth in the number of number of entrepreneurs and the amount of capital available. Certainly, the Italian entrepreneurial ecosystem was growing, with investment up 19 per cent between 2012 and 2015 according to Italia Startup. However, this was far outstripped in the growth of incubators in Italy, and more broadly – as detailed in the empirical context section – the number of start-ups and levels of investment in entrepreneurship “still lags far behind the rest of Europe” (The Local, 2015). Thus, in Phase 3 we found that pressures for local-level legitimacy remain, but they evolve and are exerted differently. Specifically, H-Farm’s challenge with respect to local level-legitimacy was no longer about explaining and justifying H-Farm’s goals and practices, because the incubator as a form – a distinct category of organizations – was now quite well understood. Rather, in this phase the challenge of local legitimacy was rooted in the need to be optimally distinct – similar but different – vis-à-vis a growing number of local rivals. Perceived Pressures for Category-level Legitimacy The category-level legitimacy pressures also evolved in Phase 3. The entrepreneurs’ desire to build legitimacy amongst category-level stakeholders – international entrepreneurs, international investors, and international accelerators – remained undimmed: in light of continued concerns about the local entrepreneurial ecosystem, H-Farm believed that access to international capital and talent was more crucial than ever. The question was how best to attract it. The story of Phase 2 was of H-Farm moving closer to what the leading US incubators were doing to build awareness of the venture among category-level stakeholders. Important progress had been made in this respect. For example, a senior manager at H-Farm, Tim O’Connell, became part of the leadership team at the GAN, a powerful signal of H-Farm’s legitimacy. And in 2012, H-Farm finally managed to secure a deal with a high profile British VC to acquire one of its start-ups, which our informants described as a landmark event in the firm’s history. However, the hoped for level of international investment had not materialized, and H-Farm was still struggling to attract a critical mass of international entrepreneurs. At the same time, there were serious concerns about whether H-Farm was viable as a US type incubator in Italy. At the EAN meetings, doubt was continually cast upon the applicability of the US model in Europe outside of the main hubs of London and Berlin. For example, a report published from the European Accelerators summit noted that, despite progress, European incubation “is still struggling to become a sustainable business” (Cruz, 2016: 9). Even in the top incubators in the US, which have access to the “best” talent and most developed ecosystem “only 2%... have a successful exit yet…, big exits can take up to 10 years – and during that time programs will have to continue funding new cohorts of start-ups, at a cost of a few hundred thousand dollars a year” (Solomon, 2015). A senior H-Farm partner put it more bluntly: “The incubator-accelerator model isn’t working outside of Silicon Valley – in fact outside of Y Combinator or Techstars… if by working you mean making serious returns” (ID48). Thus while H-Farm believed that it needed to increase its legitimacy amongst category-level stakeholders in order to attract resources from outside of Italy, it came to the view that the strategy of conforming ever more closely to US model of incubators was not viable from a commercial standpoint: such a model was ill-suited to the local context, and H-Farm risked been seen a pale imitation of the original. Given the rapid growth of incubators worldwide, H-Farm’s leadership reasoned that it would be unlikely to persuade entrepreneurs to locate there or investors to provide capital if it continued on this path. Rather, H-Farm believed that it needed to be optimally distinct at the category level. Dual Optimal Distinctiveness Work We have seen how the nature of the legitimacy pressures experienced by H-Farm evolved. At the local-level there were now multiple private incubators in Italy and so H-Farm needed not only to explain the venture to local-level stakeholders, but also to differentiate itself from local rivals. At the category-level, the explosion of incubators across the world meant that H-Farm not only needed to build recognition amongst category-level stakeholders, but also to differentiate itself from international rivals. The confluence of these pressures posed a fundamental challenge with respect to the balance between conformity and differentiation: H-farm needed to be optimally distinct at the local- and the category- levels to compete for local and international investment and entrepreneurial talent. In responding to this double optimal distinctiveness hurdle, our analysis suggests that H-Farm engaged in a process that we term dual optimal distinctiveness work. This involved: 1) practice work to develop a particular configuration of practices – a new model of incubation – which H-Farm believed was unique at the local- and the category-level, and 2) meaning work that leveraged the symbolic actions used previously to authenticate the venture at the local-level to position it as distinct at the category-level, while simultaneously leveraging the symbolic actions used previously to authenticate the venture at the category-level to position it as distinct at the local-level. Practice Work. In seeking to attain dual optimal distinctiveness, H-Farm tried to assemble a set of practices that were distinctive in the eyes of local- and category-level stakeholders. As our informants pointed out, this was something of a balancing act. If the practices that H-Farm deployed were considered too leftfield, the risk was that H-Farm would be viewed as something other than an incubator – an imposter that was not a ‘real’ example of the form. On the other hand, if H-Farm moved too close to the ‘standard’ incubator model, the risk was that it would be seen as just another ‘run-of-the-mill’ incubator, which would make it more difficult to attract resources at the local- or the category-level. To address this issue, H-Farm made some pervasive changes to its business model. Specifically, in 2015 it expanded beyond incubation into digital consulting with established businesses. At the same time, it grew significantly its educational offerings in the fields of digital technology and entrepreneurship through its “Digital Academia”. In part, the motivation was commercial because as a standalone incubator H-Farm was under pressure financially. As one informant put it, “now the acceleration programme is paid for by the corporates [i.e., through digital consulting], so there is less pressure in generating the exits” (ID49). However, the founders also saw an opportunity to transform their approach to incubation, leveraging the connections between the other two parts of H-Farm. In particular, the acceleration model that they had developed changed radically – alongside their standalone acceleration program, they began to focus on the acceleration space at the intersection of corporates and start-ups. This took a number of forms, but perhaps the most significant was a series of “Industry Accelerator” programs designed “for facilitating the connection between startups and enterprises within a specific industry” (H-Farm, 2017). These tended to focus on sectors that were deemed quintessentially Italian. For example, H-Farm partnered with Cisco to create a Food Accelerator Program, which is “dedicated to early-stage startups that are developing innovative solutions for disrupting the food and agriculture sector” (H-Farm web capture, 2017). Intriguingly, it also created a Marco Polo Accelerator Program in partnership with Chinese firm QWOS, and which is “dedicated to Italian innovative start-ups that want to hit the Chinese market” (H-Farm web capture, 2017). A number of incubators around the world are starting to partner with companies – H-Farm is not unique in this respect. Nonetheless, there were aspects of its approach that H-Farm believed made it stand out, such as its specific focus on sectors linked to Italian design and heritage. In addition, startuppers in H-Farms corporate accelerator programs had the opportunity to work with, and learn from, multiple industry partners – entrepreneurs are not tied to a single partner like in many programs. Moreover, and somewhat unusually, the programs at H-Farm were very unstructured. According to a senior manager at H-Farm: “For the start-ups, it's a little bit more difficult because most start-ups in the beginning… they are used to one model and thinking okay I come to your program so I can get financing and then move on… So I…say no it is absolutely [for] you... to drive your business and what you do with each of those brands [corporate partners] is dependent on you… we are not telling them, ‘hey now you've come here you know these guys own you’. Its [the] exact opposite… you come here and you… decide how you create value together” (ID51). Through the corporate accelerator programs, H-Farm finally managed to attract a large number of international entrepreneurs. Indeed, Italians were very much in the minority on the corporate programs: “Actually most of our start-ups now… [on] our current program [there are] 11 teams and only 2 are Italian. The rest are all from all over the world: Australia, South Africa, Germany, UK, US. I mean we've had teams from all over the world” (ID51). Moreover, while VC investment in H-Farm remained relatively limited, through its relationship with international companies it succeeded in bringing in significant investment from overseas. Crucially, in 2015 H-Farm entered a strategic partnership with British VC company InReach, which was headed by an Italian expatriate called Roberto Bonanzinga. As part of the deal H-Farm took a 20 per cent stake in the UK firm, with Bonanzinga assuming the leadership of H-Farm’s investment division. This helped attract both international capital and talent to H-Farm. Towards the end of our data collection, H-Farm announced another significant development which it believed would further allow it to be optimally distinct at both the local- and category-levels. Specifically, it received €110 million investment to build “H-Campus” – a “human thinking park” designed to serve as “a place of sharing where students, young startuppers, professors and entrepreneurs live side by side… [and] exchange ideas … on the future… and on how digital transformation will change society” (H-Farm web capture, 2017). The intention was to bring the three components of H-Farms business model – its accelerator programs, consulting, and educational initiatives – under one roof, which it believed was a “unique opportunity” (ID49) to build a set of practices that “leverages on three divisions with strong synergies and interconnections” (H-Farm, 2017) to create an “innovation platform” for digital. Meaning Work. Having developed a radically altered set of practices – which H-Farm claimed was its own unique model for incubation – H-Farm positioned itself symbolically as optimally distinct at both the local- and the category-levels. Intriguingly, in doing so it inverted its authentication work described in the previous phases: H-Farm sought to leverage what was distinctive about the Italian context to build optimal distinctiveness amongst category-level stakeholders, and it sought use its standing in the international incubation community to build optimal distinctiveness among local-level stakeholders. In doing so, H-Farm claimed both membership of, and differentiation from, the target form, at the local- and category-levels. Considering first H-Farm’s efforts to be optimally distinct at the category-level, we observed that, in addition to emphasizing its membership of the global community of incubators, the distinguishing features of the H-Farm model and local context began to be promoted to international stakeholders as a reason to confer resources to it. Evidence for this insight came from data we collected on H-Farm’s intense international networking, keynote speeches to international audiences, as well as interviews and articles published in international English-language media outlets. Through these communication channels, H-Farm sought to convey how its Italian roots helped to create a distinct philosophy and approach to innovation. For example, Riccardo tried to explain the role of Italian humanism to international audiences, as illustrated in the following quotation from an interview he gave with TechCrunch in 2016: In a time when everything was electronic – e-commence, eBay, and even my own company was called E-Tree – I wanted to create something that put humans at its core (Mari, 2016a). Crucially, the founders believed that the material features of H-Farm’s location could be used to make the venture “stand out” (ID09) in order to attract international partners. For example, in an interview with Canadian newspaper the Globe and Mail, Riccardo said the location of H-Farm was chosen: [F]irst because it is beautiful… And then, because we work with technology and innovation, it is important to have continuous contact with nature and its natural cycle (Atkinson, 2016).  More broadly, the physical qualities of H-Farm, which Wired magazine described as “the world’s most aesthetically pleasing incubator” (Rowan, 2014), were considered “part of the value of the initiative” (ID05) because it increased the credibility of the venture and its