Published online by Cambridge University Press: 01 February 2018
Paradoxical arguments and mixed empirical evidence coexist in the current literature concerning the relationship between team familiarity and team innovation. To resolve this contradiction, we apply habitual routines theory to propose that team familiarity and team innovation have an inverted U-shaped relationship. Using a data set of 68,933 R&D teams in the electrical engineering industry, our results support a nonlinear relationship between team familiarity and team innovation, and suggest that the best innovative performance is produced by moderately familiar teams. Furthermore, we find that external learning can moderate this curvilinear relationship. Theoretical contributions and future implications are discussed.