Dahlin, Eric Carl2010-10-052010-10-052010-07https://hdl.handle.net/11299/94323University of MInnesota Ph.D. dissertation. July 2010. Major: Sociology. Advisor: Professor David Knoke. 1 computer file (PDF); viii, 166 pages, appendices A-B.My dissertation takes a broad view of innovation by investigating product success among U.S. biotechnology firms across various stages of innovation including product discovery, product development, and product success. Current explanations of biotechnology product success examine one or two stages of innovation and underscore the importance of strategic alliances. However, current explanations are incomplete. First, they fail to examine whether their explanations hold across the entire innovation process. Second, estimates suggest that up to 70% of strategic alliances fail to meet their objectives (Kale and Singh 2009) and product develop remains very costly despite the high incidence of alliances in the biotechnology industry. I propose that success across the stages of innovation is associated with the scope of learning that occurs within the firm, among strategic alliance partners, and from a focal firm's network. That is, product discovery is associated with learning within the firm, product development is associated with learning among strategic alliance partners, and product success is associated with learning from the firm's overall network. While entering strategic alliances to pool resources to defray the costs of innovation is likely a necessary condition for innovation success current research overlooks the role of product development strategies. In this study I examine product development strategies that influence the likelihood of innovation success including exploration, exploitation, and ambidexterity (i.e., the simultaneous pursuit of exploration and exploitation strategies). Moreover, findings from interviews with executives in biotechnology firms provide insight into the strategies firms use to develop new drugs and evaluate them at various stages of innovation. Results from regression models support the general proposition that success at different stages of innovation varies with the scope of learning. Learning at the organizational-level (firm age and absorptive capacity) is likely to increase success at the discovery stage. Alliance partnerships are sources of learning (research alliance and development alliances) that affect product development. Network-level learning (network centrality and network experience) influences sales growth, but only for smaller firms in my sample. I also find that ambidexterity product development strategies are statistically significant predictors of success at each stage of innovation.en-USAmbidexterityBiotechnologyOrganizational learningRecombinationStrategic alliancesSociologyThesis or Dissertation