Why and How Combining Strong and Weak Ties within a Single Interorganizational R&D Collaboration Outperforms Other Collaboration Structures


  • Ingo Michelfelder,

  • Jan Kratzer

  • The authors thank the members of the Carbon Nanotube Network “Inno.CNT” for providing their time and support for this research, especially the network's Managing Directors Dr. Peter Krüger from Bayer MaterialScience AG and Dr. Holger Hoffschulz from Bayer Technology Services GmbH for their cooperation for this research as well as the management team responsible for setting up the innovative “Inno.CNT” structure which forms the basis for this research.

Address correspondence to: Ingo Michelfelder, Technische Universität Berlin, H76, 10623 Berlin, Germany. E-mail: ingo.michelfelder@tu-berlin.de. Tel.: +49 30314 21713.


The purpose of this paper is to demonstrate why and how an ambidextrous interorganizational R&D collaboration outperforms other collaboration structures in the creation of innovation. This research effort contributes to a growing stream of research in social network theory suggesting that the contradictory theories of the strength of weak ties and weak network structures on the one hand and the theory of strong ties and closed network structures on the other have a mutually reinforcing effect on innovation outcomes if combined rather than considered separately. An in-depth exploratory single case study approach within an innovatively organized national R&D collaboration allowed giving further evidence for such a superior innovation performance and for this research to contribute to theory by demonstrating why and how such a combination may lead to higher innovation output and how this effect can be actively reinforced. It is suggested that the combination of strong and weak ties should occur at the individual rather than at the project or firm level. The authors distinguish between the additive effects of the respective innovation benefits of strong and weak ties, a positive interaction effect in the portfolio of dyadic ties of an individual and a second multilevel interaction effect of weak ties embedded in the ambidextrous network structure. Referring to previous empirical findings, intellectual property regulation and structural interdependency between network members showed a higher impact than trust with regard to leveraging weak ties and are important sources for achieving the multilevel interaction effect. Managerial implications of this research are that a large network will outperform several smaller, independent networks given that the right structure and processes are in place. Direct implications for the architecture of an ambidextrous R&D collaboration are discussed, and a framework for a new form of technology R&D collaboration called “semi-open organization” is presented, which places itself between the extremes of traditional R&D in closed organizations and completely “open innovation” approaches.