Team familiarity, role experience, and performance

Much of the literature on team learning views experience as a unidimensional concept captured by the cumulative production volume or number of projects completed by a team. Implicit in this approach is the assumption that teams are stable in their membership and internal organization. In practice, however, such stability is rare, as the composition and structure of teams often changes over time or between projects. In this paper, we use detailed data from an Indian software services firm to examine how such changes may affect the accumulation of experience within, and the performance of, teams. We find that the level of team familiarity (i.e., the average number of times that each member has worked with every other member of the team) has a significant and positive effect on performance, but we observe that conventional measures of the experience of individual team members (e.g., years at the firm) are not consistently related to performance.