Dovev Lavie

Exploration and Exploitation Within and Across Organizations

Abstract Jim Marchs framework of exploration and exploitation has drawn substantial interest from scholars studying phenomena such as organizational learning, knowledge management, innovation, organizational design, and strategic alliances. This framework has become an essential lens for interpreting various behaviors and outcomes within and across organizations. Despite its straightforwardness, this framework has generated debates concerning the definition of exploration and exploitation, and their measurement, antecedents, and consequences. We critically review the growing literature on exploration and exploitation, discuss various perspectives, raise conceptual and empirical concerns, underscore challenges for further development of this literature, and provide directions for future research.

Balance Within and Across Domains: The Performance Implications of Exploration and Exploitation in Alliances

Organizational research advocates that firms balance exploration and exploitation, yet it acknowledges inherent challenges in reconciling these opposing activities. To overcome these challenges, such research suggests that firms establish organizational separation between exploring and exploiting units or engage in temporal separation whereby they oscillate between exploration and exploitation over time. Nevertheless, these approaches entail resource allocation trade-offs and conflicting organizational routines, which may undermine organizational performance as firms seek to balance exploration and exploitation within a discrete field of organizational activity (i.e., domain). We posit that firms can overcome such impediments and enhance their performance if they explore in one domain while exploiting in another. Studying the alliance portfolios of software firms, we demonstrate that firms do not typically benefit from balancing exploration and exploitation within the function domain (technology versus marketing and production alliances) and structure domain (new versus prior partners). Nevertheless, firms that balance exploration and exploitation across these domains by engaging in research and development alliances while collaborating with their prior partners, or alternatively, by forming marketing and production alliances while seeking new partners, gain in profits and market value. Moreover, we reveal that increases in firm size that exacerbate resource allocation trade-offs and routine rigidity reinforce the benefits of balance across domains and the costs of balance within domains. Our domain separation approach offers new insights into how firms can benefit from balancing exploration and exploitation. What matters is not simply whether firms balance exploration and exploitation in their alliance formation decisions but the means by which they achieve such balance.

Collaborating for Knowledge Creation and Application: The Case of Nanotechnology Research Programs

We study how collaboration and internal resources drive knowledge creation and application in university research programs. Academic collaboration with fellow university scientists drives knowledge creation, whereas collaboration with industry partners drives knowledge application. Nevertheless, contrary to prior research that has underscored the merits of collaboration, we identify an optimal level of collaboration beyond which collaboration undermines both processes. Furthermore, the availability of internal resources can either complement or substitute for collaboration depending on the level of collaboration. In particular, we find that availability of internal resources mitigates the effect of academic collaboration on knowledge creation when collaboration is moderate and complements it as collaboration becomes excessive. Thus, our study reveals the contingent value of collaboration and the interplay between internal and network resources. It enhances understanding of collaboration in nascent science-driven industries and advances the resource-based view and knowledge management research.

Knowledge Maturity and the Scientific Value of Innovations The Roles of Knowledge Distance and Adoption

How does the scientific value of innovations vary with the maturity of the knowledge that underlies them? We reconcile conflicting views in the innovation literature by introducing a contingency perspective that underscores the role of knowledge distance along technological and geographical domains. We predict an inverted U-shaped effect of knowledge maturity on the scientific value of new innovations. We further suggest that incorporating geographically distant knowledge can enhance the value contribution of knowledge maturity, whereas incorporating technologically distant knowledge or waiting for the adoption of knowledge in the industry mitigates this value. Our analysis of 5,575 biotechnology patented innovations offers support for our conjectures. We thus advance research on knowledge management and innovation by underscoring the temporal aspect of innovation and its interplay with technological and geographical distances.

The case for a process theory of resource accumulation and deployment

The resource-based view (RBV) is a useful framework for studying strategic organization. However, it focuses on rent-generating properties of resources, with less attention paid to the processes by which firms accumulate and deploy resources. I contend that organizational design that supports these processes should be integral to RBV theorizing. Despite recent debates concerning the RBV (Barney, 2001; Lado et al., 2006; Priem and Butler, 2001), its conceptualization of the firm as a bundle of resources has remained immune to criticism. This conceptualization has led scholars to consider everything that is embedded in the firm as a resource: ‘all assets, capabilities, organizational processes, firm attributes, information, knowledge, etc.’ (Barney, 1991: 101). Thus, it has obscured the distinction between resources and other organizational properties that govern resource accumulation and deployment. Related research streams shed little light on these processes. The knowledge-based view underscores some properties of knowledge (Kogut and Zander, 1992). Although several scholars distinguish flows from stocks of knowledge (e.g., Decarolis and Deeds, 1999), this literature disregards organizational constraints that limit knowledge accumulation and deployment. In turn, the absorptive capacity literature studies internalization of external knowledge, referring to the cycle of acquisition, assimilation, transformation, and exploitation (Zahra and George, 2002). It concentrates on abilities rather than on supportive mechanisms. The capabilities literature examines a firm’s abilities to achieve objectives by integrating, combining, and leveraging resources (Hoopes and Madsen, 2008). It sheds more light on the means by which firms acquire and release resources, yet falls short of revealing how firms manage their resource bases. Finally, the dynamic capabilities literature concerns the ability to reconfigure capabilities in view of path dependencies and changing market conditions (Teece et al., 1997). Capabilities can follow predetermined lifecycles (Helfat and Peteraf, 2003) or be transformed and substituted (Lavie, 2006), yet the mechanisms that support their reconfiguration have not been specified. These streams of research contribute little to the understanding of dynamic resource accumulation and deployment. Most approaches do not distinguish resource flows from stocks; some consider only one type of resource, while others underscore either internal or external resources. They neither distinguish resources from capabilities and other organizational properties nor do they explain their interplay. They associate competitive advantage with the properties of resources and 452822 SOQ10310.1177/1476127012452822LavieStrategic Organization 2012

Revisiting James March (1991): Whither exploration and exploitation?:

We revisit March’s seminal 1991 article, “Exploration and exploitation in organizational learning”, and analyze the impact it has had on scholarly thinking, providing a comprehensive and structured review of the extensive and diverse research inspired by this publication. We show that although this influence has changed significantly over the years, there are still unexplored opportunities left by this seminal work. Our approach enables us to identify promising directions for future research that reinforce the themes anchored in March’s article. In particular, we call for reconnecting current research to the behavioral roots of this article and uncovering the microfoundations of exploration and exploitation. Our analysis further identifies opportunities for integrating this framework with resource-based theories and considering how exploration and exploitation can be sourced and integrated within and across organizational boundaries. Finally, our analysis reveals prospects for extending the notions of exploration and exploitation to new domains, but we caution that such domains should be clearly delineated. We conclude with a call for further research on the antecedents of exploration and exploitation and for studying their underexplored dimensions.

Resource Reconfiguration: Learning from Performance Feedback

Abstract Resource reconfiguration enables firms to adapt in dynamic environments by supplementing, removing, recombining, or redeploying resources. Whereas prior research has underscored the merits of resource reconfiguration and the modes for implementing it, little is known about the antecedents of this practice. According to prior research, under given industry conditions, resource reconfiguration is prompted by a firm’s corporate strategy and by characteristics of its knowledge assets. We complement this research by identifying learning from performance feedback as a fundamental driver of resource reconfiguration. We claim that performance decline relative to aspiration motivates the firm’s investment in knowledge reconfiguration, and that this investment is reinforced by the munificence of complementary resources in its industry, although uncertainty about the availability of such resources limits that investment. Testing our conjectures with a sample of 248 electronics firms during the period 1993–2001, we reveal a clear distinction between exploitative reconfiguration, which combines existing knowledge elements, and exploratory reconfiguration, which incorporates new knowledge elements. We demonstrate that performance decline relative to aspiration motivates a shift from exploitative reconfiguration to exploratory reconfiguration. Moreover, munificence of complementary resources mitigates the tradeoff between exploratory and exploitative reconfigurations, whereas uncertainty weakens the motivation to engage in both types of reconfiguration, despite the performance gap. Nevertheless, codeployment, which extends the deployment of knowledge assets to additional domains, is more susceptible to uncertainty than redeployment, which withdraws those assets from their original domain and reallocates them to new domains. Our study contributes to emerging research on resource reconfiguration, extends the literature on learning from performance feedback, and advances research on balancing exploration and exploitation.

A Quest in Time: How the Maturity, Distance, and Diffusion of Knowledge Affect Innovation

Prior research has offered mixed views on the implications of knowledge maturity for the value of innovation. We seek to reconcile these views by claiming that the effect of knowledge maturity is contingent on the origin of knowledge and the extent of its diffusion in the industry. We predict an inverted U-shaped effect of knowledge maturity on the value of new innovations. Moreover, when inventors incorporate geographically distant knowledge in their innovations, we expect the value of knowledge maturity to be enhanced. In turn, incorporating technologically distant knowledge or waiting for knowledge to become diffused in the industry is likely to mitigate the value of knowledge maturity. Analysis of the citation patterns associated with 5,575 biotechnology patents of firms operating in the United States between 1985 and 2002 offers support to our conjectures. By underscoring the contingent value of knowledge maturity, our study advances innovation research and contributes to the learning literature.