Open innovation: status quo and quo vadis - an analysis of a research field

The first step in a bibliometric analysis is to collect raw data from which the necessary metadata (e.g., authors, number of citations, countries) can be obtained (Carvalho et al. 2013). Several bibliometric databases exist. However, the two largest worldwide-used databases are the Web of Science (WoS) from Clarivate Analytics and Scopus from Elsevier. The WoS contains over 90 million documents from more than 15,000 journals, while Scopus indexes around 69 million records from more than 20,000 active titles (i.e., peer-reviewed journals, books, and conference proceedings) (Forliano et al. 2021). In this study, similarly to bibliometric approaches performed in business and management studies by other scholars (e.g., Forliano et al. 2021; Secinaro & Calandra 2020; Vallaster et al. 2019; Wei et al. 2021), we consulted the WoS Core Collection for conducting the bibliometric analysis. WoS is, in fact, considered a reliable database for data mining and has become one of the primary databases used by scholars for conducting bibliometric analyses (Thelwall 2008; Waltman & van Eck 2012). In line with previous influential bibliometric studies on the same topic (e.g., Randhawa et al. 2016; West & Bogers 2014), we used the search phrase “open innovation” in title, abstract, or keywords. The search was conducted in January 2022. The initial search returned 7534 records. Then, restrictions on the year (from 2010 to 2021 only), document type (articles only, excluding for instance book chapters and proceedings), language (English only), and subject (Business and Management only) were applied, refining the sample to 1998 articles. We also further limited our review to studies in journals ranked 2, 3, 4, or 4* in the Academic Journal Guide/ABS 2021 journals list, as including only top-tier academic journals remains the most frequently adopted method for identifying scholarly debates and research trends when performing the review of specific topic-related literature (Atewologun et al. 2017). The Academic Journal Guide/ABS is not the only journal ranking, but is considered among the major ones (Kraus et al. 2020). Several articles in the field of business and management studies have used this ranking to further refine the search (e.g., Bertello et al. 2022c; Vrontis & Christofi, 2021). Others have used different criteria such as the JCR Impact Factors (IF) by Clarivate Analytics (e.g., Kunisch et al. 2015; Überbacher 2014). Moreover, these different rankings can be compared and converted (Bouncken et al. 2015; Kraus et al. 2020). Based on Kraus et al.’s (2020) conversion table of leading academic journal rankings, a journal ranked ≥ 2 in the Academic Journal Guide/ABS is comparable to a journal ranked ≥ C in the VHB Jourqual (JQ) and ≥ 1.5 in the JCR Impact Factors (IF) by Clarivate Analytics. Based on this information, this criterion allowed us to include in our research the most relevant state-of-the-art knowledge of the topic of interest while keeping the sample manageable. Based on these exclusion/inclusion criteria, our final result was 1515 records. Based on our purpose to explore whether and to what extent the open innovation area of studies is evolving in response to the emerging transition towards a more digital and sustainable society, we set 2015 as a threshold to compare two different time periods (2010‒2015 and 2016‒2021). We followed two criteria to identify the year to be selected as a threshold. We included in our criteria considerations about the impact of digitalization and sustainability issues on scientific production. First, we conducted a preliminary search on the WoS database to explore the research trend of digital transformation studies. We indeed selected the term “digital transformation” and we looked at the distribution of papers per year. The year that shows the greatest increment percentage is 2016, where the number of papers increased from 32 to 146 (+ 456%). 2015, however, represents a turning point also as concerns sustainability issues. In 2015, the United Nations released the Sustainable Development Goals (SDGs), providing a list of 17 goals such as “no poverty” (SDG 1), “good health and well-being” (SDG 3), “decent work and economic growth” (SDG 8), articulated in 169 targets to be achieved by 2030. The SDGs were immediately acknowledged as the translation of grand challenges, raising the awareness of practitioners and scholars that the progress required to achieve these goals involves collective, collaborative, and coordinated efforts (George et al. 2016). They thus represent a relevant turning point to explore potential shifts in business and management scholarly literature (Ricciardi et al. 2021).

Our paper sample is characterized by 437 papers within the time period 2010‒2015 and 1078 papers within the time period 2016‒2021 (see Fig. 1 for a visual representation of the data extraction activity). Part of our descriptive analysis, as well as the keyword co-occurrence analysis and the content analysis, were structured to compare these two different time periods.

The analysis shows four different clusters in the time period 2010‒2015 (Fig. 4). The red cluster (33 items) is more focused on co-creation processes and the figure of users as innovators. The green cluster (30 items) mainly refers to the inbound model of open innovation. The blue cluster (22 items) is focused on knowledge management and governance issues. Finally, the yellow cluster (21 items) includes keywords related to the strategic aspects of open innovation.

The analysis shows five different clusters in the time period 2016‒2021 (Fig. 5). The red cluster (33 items) outlines relevant opportunities and challenges in open innovation research. The green cluster (30 items) focuses on the human side of open innovation. Terms from the blue cluster (29 items) mainly relate to university-industry-government collaboration. The yellow cluster (21 items) refers to tensions and paradoxes in open innovation research. Finally, the purple cluster (16 items) introduces terms at the intersection between sustainability and open innovation.

The network analysis of the 15 most cited empirical studies in the time period 2010‒2015 (see Table 1 for the full list of papers included) showed the predominance of four theories/theoretical approaches (i.e., organizational learning theories, Schumpeterian theory of entrepreneurship and innovation, resource-based view, and knowledge-based theory). Another four theories (i.e., attention-based theory, evolutionary theory, transaction cost economics, and relational view) are instead mentioned only once. We also identified, although they are not explicitly depicted in the figure, two bridging concepts (dynamic capabilities and absorptive capacity). These two theoretical concepts have significantly informed open innovation research, linking numerous of these theories (see Fig. 6).

Mentioned in 4 out of the 15 most cited empirical studies in the time period 2010‒2015 (Berchicchi 2013; Brunswicker & Vanhaverbeke 2015; Laursen & Salter 2014; Mina et al. 2014), the resource-based view is one of the most influential theories of the firm in the field of strategic management. According to this perspective, a firm requires a unique collection of resources, competencies, and capabilities to be competitive (Barney 1991; Wernerfelt 1984). To create a competitive advantage and capture above-normal rates of returns (i.e. rents), these resources must, by definition be scarce, valuable, difficult-to-imitate, and reasonably durable (Barney 1991; 2001). In other words, it is the result of the use of resources and capabilities that are owned and controlled within the boundaries of a single firm that explains the differences in performance between firms (Bierly and Chakrabarti 1996). This view about how companies develop and sustain a competitive advantage, however, is typically in line with the closed innovation framework. So, why is the resource-based view one of the most advocated theories to study open innovation? A partial reason may be that both the resource-based view and the open innovation model emphasize the importance of resources and competencies to achieve sustained competitive advantage (Vanhaverbeke and Cloodt 2014). However, how can a perspective that stresses independence and the crucial role of internal resources and capabilities with the open innovation perspective of exploiting the interdependences between complementary resources be aligned? The existing literature has suggested two ways: first, by combining synergistically the resource-based view with other theories; and second, extending it. A synergic combination can be found in the relational view of the firm. The relational view emphasizes that critical resources can and should also be found outside the firm’s boundaries (Dyer and Singh 1998). Collaborating firms that combine resources in unique ways may realize a competitive advantage over others that compete on the basis of a stand-alone strategy. The relational view identifies complementary resources or capabilities of firms as a potential source of inter-organizational competitive advantage: this is in line with a major premise of open innovation to consider the sourcing of knowledge from external partners a source of competitive advantage (Alexy et al. 2018). The relational view of the firm can therefore also complement the knowledge-based theory. The main assumption behind the knowledge-based view is, in fact, that the most important strategic resource for sustained competitive advantage within organizations in a knowledge-based economy is the knowledge owned by companies.

Attempts to extend the resource-based view and knowledge-based theory can benefit from the theoretical concept of dynamic capabilities. By relying on internal capabilities only, companies are likely to get stuck in “core rigidities” (Leonard-Barton 1992) or the so-called familiarity trap (Ahuja & Lampert 2001), reducing the chances that firms take advantage of new technological opportunities (Lunn and Martin 1986; Levin et al. 1985). Teece has therefore introduced the concept of dynamic capabilities (Teece 2007), referring to firms’ ability to expand their narrow search horizon and combine internal and external knowledge that originates in the core as well as the periphery of their business ecosystem. According to Teece, firms can overcome the limits of internal learning by making use of sensing, seizing, and reconfiguring capabilities. More specifically, sensing capabilities support the discovery and evaluation of valuable external knowledge, seizing capabilities sustain adaptation and integration processes enabled by internal investments in R&D, good governance mechanisms, and the implementation of lateral and vertical communication, while reconfiguring capabilities are particularly relevant to realign the organization to integrate external knowledge and to develop a culture that promotes collaboration (Bogers et al. 2019).

Figure 6 shows how another theory that is particularly recurrent in open innovation works in the time period 2010‒2015 is the Schumpeterian view of entrepreneurship and innovation. Central to our understanding of the innovation process is how organizations search for knowledge (Nelson and Winter 1982). These search processes are described by Schumpeter as characterized by the need for firms to search for and carry out “new combinations” of technologies, knowledge, and markets with the purpose of producing other things, or the same things through different methods (Schumpeter 1982).

In order to aid our understanding of the Schumpeterian view from an open innovation perspective, the concept of dynamic capabilities may be helpful anew. Teece et al. (1997) suggest that dynamic capabilities have the capacity to reconfigure, redirect, transform, shape, and integrate central knowledge, external resources, and strategic and complementary assets. They will allow the firm to respond to the challenges presented by the Schumpeterian ever-changing world, made of competition and imitation (Teece et al. 1997). The Schumpeterian view of entrepreneurship and innovation is linked to the resource-based view but also (dotted line) to the knowledge-based view. According to the Schumpeterian rent creation logic, in fact, knowledge and its recombination can be considered a core intangible asset that contributes to building sustained competitive advantage.

Another important theoretical approach that emerges in our network analysis relates to organizational learning theories. Although organizational learning is often seen as a field of study rather than a theory, we have included it in our network analysis since its use in the field of open innovation draws on learning theories (e.g., Bower & Hilgard 1981; Harlow 1949). Organizational learning can be defined as a change in the organization’s knowledge that occurs as a function of experience (Argote and Miron-Spektor 2011; Fiol and Lyles 1985). The knowledge the organization develops can be explicit or it can be tacit and difficult-to-articulate (Kogut and Zander 1992), and it can also be converted from one type to another (Nonaka and Von Krogh 2009). The knowledge can manifest itself in a variety of ways, including changes in cognitions, routines, and behaviors. Organizational learning is therefore linked to the knowledge-based view. A firm’s sustained competitive advantage depends, in fact, on both the firm’s ability to embed knowledge in a variety of repositories or knowledge reservoirs, including tools, routines, social networks, and transactive memory systems (Argote and Ingram 2000; Walsh and Ungson 1991), but it also resonates well with the open innovation principles, since organizational learning often depends on the combination of internal and external knowledge that originates in the core as well as the periphery of their business ecosystem. Organizational learning is often mediated by the concept of absorptive capacity, another bridging concept for the theories that support a better understanding of open innovation. Absorptive capacity suggests, by definition, that higher levels of absorptive capacity lead to firms’ better understanding of the knowledge received, making it easier to unlock and capture the intrinsic value of such knowledge and apply it for commercial purposes (Carayannis et al. 2012). In the same way as the dynamic capabilities concept, the intrinsic dynamism of the notion of absorptive capacity is appropriate for understanding open innovation in fast-changing settings.

Another perspective that has been used to explore open innovation is the transaction cost theory. Open innovation is about setting up relations with external innovation partners. The inter-organizational relations that occur between firms reflect transactions between different legal entities. Firms can, in fact, collaborate in many different ways with their innovation partners and the choice between these different types of collaboration is at the core of the transaction costs economics (Williamson 1989). Transaction costs tend to increase when fear of opportunism and asset specificity are higher. However, collaboration can potentially cope with a high degree of asset specificity and can also lower uncertainty and opportunistic behavior by monitoring partners’ performance, balancing their contributions, and creating interactions based on mutual reciprocity and long-term relationships (Bogers 2011; Vanhaverbeke and Cloodt 2014), however, have proposed to integrate the transaction cost economics with the transaction value theory, since ‘careful observation of open innovation deals reveals that companies choose a particular type of collaboration governance to jointly maximize the [strategic] value of a transaction rather than to minimize transaction costs’ (Vanhaverbeke and Cloodt 2014, p. 263). Moreover, the link between open innovation and the theories of learning suggests that the value of cooperation in open innovation contexts is based not only on transaction costs objectives but also on learning opportunities (Kogut and Zander 1992). Transaction cost economics and the resource-based view (as well as the knowledge and the relational-based view) are fundamentally different. The former invokes a contractual view whose focus is explicit and implicit contracts between employers, employees, and other contractors. This view, however, has been criticized for neglecting non-contractual relations that affect transactions such as trust and loyalty, and for providing static explanations of organizational arrangements. On the contrary, the resource-based view places organizations in a disequilibrium environment where they develop resources and skills to achieve competitive advantage. Despite their differences, however, resource-based view and transaction cost economies also complement each other (Argyres and Zenger 2007; Mayer and Salomon 2006). In this regard, the concept of dynamic capabilities can bridge transaction cost and non-transaction cost arguments into a more comprehensive theory that can be fruitfully used to reveal the dynamics of open innovation. A focus on comparative static explanations, where one organizational arrangement is deemed to have lower (transaction) costs than another, leads to inadequate treatment or neglect of dynamic aspects of the problem, notably learning, innovation, and technological development. As Teece and Pisano (2003) argued, firms do not only deal with transaction costs but also with many types of arrangements where injecting high-powered (market-like) incentives might well be destructive of the cooperative activity and learning (Hodgson 1998).

Another theory emerging from our analysis is the attention-based view. This theory postulates three core principles: first, decision-makers’ focus of attention has an impact on strategic choices and outcomes; second, this attention is contextually situated; third, this situated attention is socially structured (Ocasio 1997). Although this theory is sparingly used in open innovation research, it can potentially gain more relevance, since the view of humans as boundedly rational and characterized by cognitive limits raises issues on how they may allocate their attention and behave in inter-organizational contexts characterized by information overload, multiple and sometimes conflicting goals, and potential opportunistic behavior (Bertello et al. 2022a; Giusti et al. 2020). An attention-based approach can provide insights in exploring how organizations absorb knowledge and how they balance exploration and exploitation activities (Miller & Martignoni 2015).

Finally, our social network analysis also includes one attempt to extend evolutionary economics. Ghisetti et al. (2015), in their Research Policy paper, investigate the effect of knowledge sourcing and absorptive capacity on firms’ environmental innovation, extending the open innovation paradigm to environmental innovation (i.e., open eco-innovation), and contributing to a growing body of literature that analyzes the potential of evolutionary economics to explain sustainable development and environmental policies.

The network analysis of the 15 most cited empirical studies in the time period 2016‒2021 (see Table 2 for the full list of papers included) showed the increased predominance of the resource-based view and knowledge-based theory of the firm (recurring 4 and 8 times, respectively), often used in combination. This result can, in part, be explained by the increasing attention (see Table 4) that the Journal of Knowledge Management, one of the leading journals in the knowledge management field, is devoting to the open innovation discourse. Another two theories (i.e., theories of organizational learning and contingency theory) are advocated twice, while a broader range of theories has been explicitly mentioned once (i.e., transaction cost economics, cognitive theory, theories of creativity, game theory, social exchange theory, human capital theory) (see Fig. 7).

A comparison between the network analysis from the two time periods shows that 10 theories have been advocated in the time period 2016‒2021, against only 8 in 2010‒2015. There was an increase also in the number of articles that have used more theories and/or theoretical approaches contemporaneously, as can be seen from the number of circles and the number of lines (22 vs. 20 and 13 vs. 12, respectively). This might suggest that scholars are gradually trying to combine existing theories to capture the complexity of open innovation processes. It can also be seen how, beside the well-grounded conversation around the resource-based view and knowledge-based theory and their use in open innovation studies, the theoretical discourse has also opened the door to micro-level theories. This reflects the recent turn in open innovation studies towards exploring the human side of open innovation (e.g., Albats et al. 2020; Bogers et al. 2018b) by employing a microfoundational approach as a theoretical lens to understand how the organizational context to which individuals are exposed and their actions and interactions influence organizational capabilities (Felin et al. 2015). Indeed, cognitive theories, theories of creativity, and human capital theories have recently been used to reveal the microfoundations of open innovation-related capabilities such as absorptive capacity, relational capabilities, and, more generally, dynamic capabilities. Instead, a macro-level focus (i.e., open innovation in relation to institutions, policies, socio-ecological systems), at least in the most influential papers in terms of citations, is still lacking,

Transaction cost economics is only mentioned in one paper, as in the previous time period. Inter-organizational relationships, however, do not include only the cost of transacting, but also cognitive and organizational costs (Cassiman and Valentini 2016). The social-exchange theory, for instance, has been increasingly adopted by knowledge management scholars to inform research on knowledge sharing (Liao 2008) and knowledge hiding (Perotti et al. 2021). Its underlying assumption is that people rely on cost-benefit analysis to determine the risks and benefits of each decision and behavior. As a consequence, they will be willing to share knowledge when the (perceived) rewards are higher than the (perceived) costs (Lin and Huang 2010; Tsai and Cheng 2012; Wu and Lee 2017). Since the quality of knowledge flows plays a key role in open innovation (Chesbrough and Bogers 2014), the social exchange theory can provide valuable insights to explore the antecedents of knowledge flow (Martinez-Conesa et al. 2017) at both firm and inter-organizational levels.

One paper in our sample (Parker and Van Alstyne 2018) has also made use of game theory as part of an extensive theory that studies optimal decision-making processes. In their paper, Parker & Van Alstyne (2018) developed a sequential innovation model that explores the optimal levels of openness and of intellectual property duration in a platform ecosystem. Based on the first trade-off they identified, closing the platform increases the sponsor’s ability to charge for access, while opening the platform increases the developer’s ability to build upon it. According to the second trade-off, the longer the third-party developers retain rights to their innovations, the higher the royalties they and the sponsor earn, but the sooner those developers’ rights expire, the sooner their innovations become a public good upon which other developers can build.

Finally, a theoretical approach that has been mentioned twice in the 15 articles analyzed in the time period 2016‒2021 is the contingency theory (Martinez-Conesa et al. 2017; Popa et al. 2017), an organizational theory developed in the second half of the twentieth century around the core argument that there is no best way to organize a corporation, to lead a company, or to make decisions. Instead, the optimal course of action is contingent (dependent) upon the internal and external situation. This theoretical approach has informed prior studies arguing that companies’ innovation strategies are contingent on both internal and external factors (Gibson and Birkinshaw 2004; Jansen et al. 2006). Accordingly, previous open innovation research suggests that firms’ migration toward opening up innovation strategies is also influenced by internal and external contexts, from technological development to institutional pressures (e.g., Bertello et al. 2022a; Huizingh 2011). In this vein, the existing literature has suggested that opening up innovation strategies is more suitable in business environments characterized by globalization, competitive intensity, and market and technological turbulence (Akgün et al. 2019; Huizingh 2011).

A collective endeavor to reconceptualize open innovation first requires a critical reflection on how to define open innovation. Based on the widely accepted definition provided by Chesbrough and Bogers (2014, p.17), open innovation is currently defined as “a distributed innovation process based on purposively managed knowledge flows across organizational boundaries, using pecuniary and non-pecuniary mechanisms in line with each organization’s business model”. Recently, Bogers et al. (2020) have introduced the concept of sustainable open innovation, integrating the previous definition of open innovation with the definition of sustainable development, as developed by the World Commission for Environment and Development report for the United Nations (WCED, 1987). Sustainable open innovation is therefore now defined by Bogers et al. (2020, p. 1507) as “a distributed innovation process which is based on purposively managed knowledge flows across organizational boundaries, using pecuniary and non-pecuniary mechanisms in line with the organization’s business model, thereby contributing to development that meets the needs of the present without compromising the ability of future generations to meet their own needs”. While their new definition responds to the need to adapt existing open innovation definitions to emerging organizational and societal needs, it raises questions about how to do so. We invite scholars to reflect on the existing definitions of open innovation to understand whether they are still valid in light of the disruptive changes that we have outlined in our paper. More specifically, the open innovation community could question whether integrated definitions such as the one provided by Bogers et al. (2020), which combine the core elements of the traditional definition of open innovation with the definition of sustainable development, reflect a sufficiently critical approach to capture and describe with rigor the phenomenon under analysis. Their definition, for instance, stresses how the outcome of open innovation activities is no longer exclusively a firm’s competitive advantage, but also the sustainability of the system in which firms are embedded (Ricciardi et al. 2021). However, the definition seems to overlook the inherent paradoxes of combining market and sustainability logics and short-term and long-term perspectives. Moreover, the ongoing efforts of researchers, practitioners, and policy makers to promote open innovation at a broader, societal level, as well as emerging technologies, such as digital platforms and the Internet of Things, pave the way to explore not only how traditional organizations adapt their structures, processes, and culture to implement open innovation activities, but also how new open innovation models that are not necessarily structured around a focal firm emerge (see Bertello et al. 2021b; Gegenhuber et al. 2023). To conclude, the open innovation research community initially developed around a limited number of scholars interested in the managerial implications of this emerging innovation paradigm. Nowadays, however, contributions to the open innovation research domain come from an increased number of scholars from different communities. As a consequence, the open innovation concept has been expanded and is being applied to a wide set of contexts. Future endeavors to define open innovation should therefore avoid, at the same time, both concept stretching, which may be caused by a unique all-encompassing definition of open innovation, and redundancy, which may result from the multiplication of definitions. (Re)newed definitions must also go hand in hand with new or (re)newed visualization tools that might support conceptualization work in open innovation studies. Therefore, we also encourage scholars to question and adapt existing open innovation visual frameworks, or to develop new ones (Vanhaverbeke 2013). For instance, is the innovation funnel still valid to align open innovation with recent calls for approaches that invoke a pluralistic and recursive view of innovation as a never-ending and constantly (re)negotiated process in which means and outcomes can be considered as mutually enabling and constituent of one another (see Farjoun et al. 2015).

Open innovation has principally developed as a field of studies very closely linked to the community of practitioners. Chesbrough himself, who coined the term, has long years of experience in the industry. Open innovation research has therefore privileged practical application over theoretical explanation. As a consequence, we have no specific open innovation theories (Gassmann and Enkel 2004,) but a wide range of applications of existing theories of the firm (Vanhaverbeke and Cloodt 2014), as confirmed by our studies, which registered a massive use of the resource-based view and the knowledge-based theory of the firm. The analysis of the second time period, however, showed that the microfoundations of open innovation are gaining attention, with scholars increasingly eager to explore not only the technical but also the human side of open innovation, as well as the cross-level nature of open innovation by revealing macro-micro-macro level interactions. Combining these insights from the network analysis of theories with the results of our keyword co-occurrence analysis, which showed us how most of the words referring to the technological and to the human side of open innovation were respectively falling into two different clusters in the time period 2016‒2021, we may conclude that we still lack a relational view of these two domains. In other words, we now have increasing insights about how the diffusion and adoption of new technologies impacts firm performance via open innovation practices. We are also gradually gaining knowledge about how human skills and organization culture influence firm openness. However, we still know little about how the technological and the human are constitutively entangled (see Leonardi 2013) and how this entanglement nurtures open innovation endeavors.

On the other hand, what we are also missing in the open innovation literature are macro-level discourses. In this regard, Bertello et al. (2022a) have recently invoked an institutional view of open innovation to consider firms’ efforts to adapt not only to technical pressures but also to societal expectations. According to their view, “the rapid spread of successful cases of open innovation adoption, and the increasing number of public policies promoting open innovation, may lead firms to address it as a way to conform to a societal mandate, or legitimacy, even though these pressures contradict internal needs for efficiency”. This reflection paves the way for exploring the diffusion of open innovation practices among less proactive organizations, as well as the unintended consequences (also called in the open innovation literature ‘dark sides’) of promoting the opening up of innovation processes.

Another theory that might help to explain theoretically open innovation is the stakeholder theory (Shaikh and Randhawa 2022). By conceptualizing firms as organizations at the center of an ecosystem of actors (i.e., stakeholders) who interact with the organizations and who are carriers of different interests, stakeholder theory invites organizations to rethink what creating value means in a complex society where business, social, and environmental concerns are intertwined (Freeman 1984). Moreover, it also suggests that this multiplicity of views and interests can be better achieved through collaborating directly with the different stakeholders (be they employees, customers, universities, governments, etc.). Indeed, recent perspectives that see stakeholders no longer as recipients but rather as creators (or co-creators) of value (Freudenreich et al. 2020), and multi-stakeholder engagement as enablers of collective action when focal firms engage in shared governance forms rather than hub-and-spoke governance models (Bridoux and Stoelhorst 2020), could inspire further reflections on how to manage innovation across distributed networks of stakeholders. Moreover, a multi-stakeholder view of collaborative innovation links well to other theoretical approaches such as the institutional logics perspective and the paradox lens. The institutional logic perspective, for instance, might suggest how to organize open innovation in terms of goal setting, decision making, roles and tasks, communication style, and monitoring in order to navigate multiple, often conflicting logics. The paradox lens, instead, could contribute to shed light on the tensions that characterize inter-organizational contexts: tensions between conflicting institutional logics, but also between competition and collaboration, knowledge transfer and knowledge protection, and many others. Moreover, some of the operational paradoxes of the pre-digital age (see Westerman et al. 2014) that can be associated with open innovation, such as standardizing vs. empowering, controlling vs. innovating, and orchestrating vs. unleashing, can now be addressed in a digitalized world through ‘both-and’ approaches rather than trade-offs based on ‘either-or’ alternatives. We invite scholars to explore both how traditional open innovation paradoxes evolve and how new paradoxes emerge within a renewed open innovation research landscape. Moreover, although open innovation research has focused predominantly on paradoxes of learning and organizing, we know little about what Smith and Lewis (2011) have described as paradoxes of performing and belonging. Paradoxes of performing can emerge, for instance, in open innovation contexts at the intersection between the individual, the organizational, and the inter-organizational level, when individuals are called to perform different, often contradictory roles, while paradoxes of belonging reflect identity tensions between the individual and the collective, as individuals and groups seek both homogeneity and distinction. A multi-stakeholder view of open innovation also stimulates reflections on power dynamics in inter-organizational contexts, especially in purpose-driven initiatives where coordination and the model of authority are often unclearly defined. The role of marginalized actors, who are often beneficiaries of innovative solutions, and the contribution of firms that are called to partner a many-to-many stakeholder rather than a firm-centric game, raises questions in terms of power relations that are worth addressing.

Also, the systemic nature of the challenges that open innovation is called to address opens the doors to new paths that invite open innovation scholars to cross-fertilize open innovation studies with other fascinating perspectives. The role of open innovation in addressing environmental problems, for instance, can be explored by theoretical approaches that frame their analysis in system theories, such as social-ecological studies (Dentoni et al. 2021; Olsson et al. 2004). More specifically, the Schumpeterian view of entrepreneurship and innovation (advocated several times in the papers included in our sample), according to which entrepreneurs’ and innovators’ fervid distributed experimentation has, per se, positive consequences for the system’s performance, does not explain how innovation switches from the engine of (system-threatening) economic growth to being the engine of sustainable development that requires the preservation and (re)generation of social and environmental resources for the collective use (Ricciardi et al. 2021). In other words, a more comprehensive view of open innovation that considers the joint innovators as part of a social-ecological system characterized by many possibly hidden and nonlinear fragilities would inform about the possible negative disruptive consequences of creatively destructive actions.

Moreover, the recent association of open innovation with collective action (Mair and Gegenhuber 2021) may invite scholars to draw on the theory of the commons (Ostrom 1990) to explore the social dilemmas that emerge between individual and community interests. Indeed, when open innovation unfolds as collective action that involves multiple actors with different institutional logics, a social dilemma is presented to the community of innovators: they might choose short-term, personal benefits rather than constructing a common logic (Ansari et al. 2013), thus increasing the risk that the long-term collective benefit will be lost. Social dilemmas when open innovation mobilizes collective action can also be explained by drawing insights from game theory, another theory that is mentioned by the papers in our sample. Game theory can also prove to be useful to explore coopetition dynamics in open innovation. In this regard, Le Roy et al. (2018) contend that, together with network theory, resource-based theory, and capabilities-based theory, game theory contributes to an optimistic perspective or view of competition. In fact, one of the core postulates of game theory is that the best partner in a business is a competitor, which implies that both collaboration and competition with such a partner provide a suitable positive-sum game (Heiets et al. 2021).

To conclude, three reflections arise from our review in terms of theoretical developments in the field of open innovation studies. First, theories of the firm are still valuable, but not sufficient to explain open innovation in the face of the turmoil caused by ongoing transitions towards more digital and sustainable organizational and inter-organizational practices. Second, open innovation research would benefit from the combination of existing management theories, especially across levels. Finally, the third reflection is basically an open question: will existing theories be enough to explain open innovation in the future or should we focus our efforts on grounding new theorizing?

Open innovation has been researched from a variety of methods and techniques. In the early stages of open innovation research, case studies were the predominant method because of their usefulness in illustrating successful cases and informing practitioners about best managerial practices (e.g., Chesbrough & Crowther 2006; Chesbrough 2007). Over the years, this method has been successfully reproposed to illustrate the role of open innovation in new domains such as sustainability (e.g., Bogers et al. 2020). Recently, research has also made use of case studies to shift the focus to the dark sides of open innovation, shedding light on the challenges and the possible drawbacks of open innovation (e.g., Bertello et al. 2021a; Stefan et al. 2022). Multiple case studies have been conducted likewise to carry on analyses between successful and unsuccessful cases (e.g., Grama-Vigouroux et al. 2020; Lazzarotti et al., 2013).

On the other hand, quantitative analyses have mainly focused on exploring the relationships between open innovation and firm performance (e.g., Ahn et al. 2015; Bianchi et al., 2016; Du et al. 2014; Greco et al. 2016), as well as the antecedents of open innovation at both the individual and organizational level (e.g., Bogers et al. 2018b; Naqshbandi & Tabche 2018). These studies either draw their hypotheses from close domains (e.g., inter-organizational collaboration, innovation projects) and apply them to open innovation empirical contexts, or develop research designs to test insights from open innovation case studies and conceptual papers. However, the most employed quantitative approaches in open innovation studies rely on conventional statistical (symmetrical) approaches such as regressions (e.g., Bogers et al. 2018b; Mention 2011; Papa et al. 2018) and structural equation modeling (e.g., Huang et al. 2013; Santoro et al. 2018) that may not adequately explain non-linear and complex real-world business phenomena. Linear statistical analyses with finite contextual factors could indeed suffer from prediction inaccuracy (Pappas and Woodside 2021). This prediction inaccuracy is significantly increased by digital and sustainability transformation processes that cause shifts in market circumstances and environmental conditions. Recently, organizational and management research has invoked the use of asymmetrical techniques that draw on the ontological assumptions of complexity theory, such as fuzzy-set qualitative comparative analysis (fsQCA) (Kumar et al. 2022). FsQCA is a statistically-informed configurational approach that aims to bridge and transcend the qualitative‒quantitative divide in social sciences (Ragin 2014). Research on open innovation still counts sparse contributions that have used complexity theory and fsQCA. This method could, however, be used increasingly in the future to depict the combinations of conditions that lead to the absence of an outcome, be it positive or negative. For instance, Peris-Ortiz et al. (2018) have used this method to explore the combination of factors that enable incremental innovation and foster radical innovation in open innovation contexts, focusing principally on human resources and organizational learning capabilities, while Kumar et al. (2022) have suggested applying fsQCA in open innovation studies to explore coopetition, trust and distrust dynamics, and human capital.

Changes in the methodologies adopted should also consider how open innovation practices are changing. For instance, the diffusion of digital platforms and online communities actually challenges traditional ways of collecting data and conducting field work. In this regard, the keyword netnography has emerged from the keyword co-occurrence analysis for the time period 2016‒2021. This emerging qualitative research method is already widely used in marketing studies (e.g., Pera et al. 2021). The term combines the words ‘internet’ or ‘network’ with ‘ethnography’ to highlight the role of online participant observation in exploring closely the dynamics of social interaction in contemporary digital communications contexts (Kozinets 2015). Its use could assume increasing relevance in exploring the role of users and customers in shaping open innovation practices such as new product development in an ever-increasing digitalized world (Bartl et al. 2016). Netnography, and, more generally, digital ethnography, could also represent a valuable tool for scholars who want to either extend offline data collection or explore social interaction dynamics in online contexts (Pink et al. 2015). In particular, this research method can be relevant to research online participatory contexts such as online hackathons and, more generally, crowdsourcing initiatives. These events, rapidly spread during the COVID-19 crisis, seem to be destined to become enduring models of innovation, even after the end of the pandemic, due to their capacity to gather at the same time different people from around the world (Bertello et al. 2021b). Ethnographies, be they digital or in person, would also contribute to exploring the temporal unfolding of open innovation activities, supporting a longitudinal perspective that is often lacking in open innovation studies. Indeed, open innovation studies have often prioritized variance theories, which refer to conceptual constructions that relate variables to one another, thus reducing time to a lag effect, or compressing it into variables, while process-based theorizing (i.e., ‘conceptual constructions that focus on the way in which phenomena emerge, evolve or terminate over time through activities and events’, see Cloutier & Langley 2020, p.3) is under-developed (Alam et al. 2022). Idea generation contexts such as hackathons, for instance, could be explored by paying attention to the subsequent phases of the innovation journey to understand how hackathon ideas can scale up over time and generate long-term impact (Mair and Gegenhuber 2021), while a linear conception of leveraging external sources as originally conceived in the early stages of open innovation research (West and Bogers 2014) might be further overcome by recursive process models that consider feedback loops and reciprocal interactions. Similarly, open innovation projects that require organizational change and digital transformation might be explored by paying attention to how different organizational structures, practices, and culture evolve and influence each other over time. Indeed, a process turn in open innovation studies might provide scholars with new conceptual tools to capture the complexity of an open innovation journey or an organizational re-design by considering chains of interconnected events, activities, and/or interactions in which the preceding events, activities, and/or interactions enable and/or constrain the following ones (see Langley et al. 2013).

Finally, in order to capture the complexity of open innovation dynamics in the face of sustainability challenges that are systemic by nature (Ferraro et al. 2015), we invite scholars to study open innovation practices by adopting the analytical technique of zooming in and zooming out, originally developed by Nicolini (2009) and, more recently, applied to sustainability and grand challenges studies by other scholars (see Jarzabkowski et al. 2019; Schad & Bansal 2018). By zooming into the organizing aspects of open innovation practices and then zooming out to the wider consequences of these practices, scholars would be able to capture the system-level consequences of open innovation in the face of grand challenges and to arrange organizational and inter-organizational practices accordingly. Another useful modeling tool to explore the system-level consequences of open innovation activities is system dynamics (see Richardson 2011). System dynamics is a valuable approach to understand the nonlinear behavior of complex systems and, therefore, to assess the social and environmental impact of open innovation practices over time by making use of stocks, flows, internal feedback loops, table functions, and time delays. Forliano et al. (2020), for instance, used this method to investigate the credit collection process of a municipally-owned company. By leveraging a system dynamic model, the authors moved beyond the traditional logic of linearity to capture the systemic, unintended, and delayed implications of decision-making activities related to the provision of public services and consider citizens as active stakeholders in the process of public value co-creation.

To stimulate a turn into practice-based and process-based approaches in open innovation studies, we draw on Weick’s intuition of translating nouns into verbs in organization studies (Weick 1969), and we introduce the concept of open innovating, inviting open innovation scholars to explore related phenomena by prioritizing processes over outcomes and activities over products.