A Generalist Activity Paradox: Individual Performance and Collective Production Efficiency on Open Innovation Platforms

This study examines the relationship between generalist activity and multilevel performance on open innovation platforms for new product development. In such settings, task decomposition allows contributors to engage flexibly across multiple task types, giving rise to generalist activity, defined as the distribution of engagement across a wide variety of task types. Drawing on the generalist literature and the theory of collective production, we theorize how generalist activity relates to both individual and collective performance of crowds. Our multilevel framework and empirical findings indicate that generalist activity is positively associated with individual performance. However, due to the emergent nature of collaboration, a higher level of generalist activity within a crowd is associated with lower collective production efficiency, revealing a multilevel tension. This tension is attenuated as the crowd size increases. Mechanism analysis further shows that valuable contributions are more likely to emerge when contributors with generalist activity build upon task-focused inputs from preceding contributors with low levels of generalist activity. Overall, this study provides theoretical implications for open innovation platform research, generalist literature, and the theory of collective production.