This paper examines the variables affecting innovation processes performance through the analysis of 315 papers and books in management, engineering, software development, and design and architecture. Through an integrative literature review, and inductive research process, we cluster research findings around common contributions, and synthesize them into fifty six different schemas for explaining innovation performance. We later clustered these schemas into twelve major dimensions affecting innovation process performance. Three of these dimensions relate to firm-level strategic context; eight are associated to within-process, operational or task-related factors; and one responds to team dynamics. As result, we propose a construct that relates these dimensions and their associated variables, and propose an explanation of the mechanisms by which they affect innovation process performance. Across the management literature, innovation process performance is explained as resulting from strategic factors and senior management leadership; the technical capabilities and creative potential of individuals, and communications within team members; and process formalization and task partitioning. We find a beginning acknowledgement in the relevance of social cognition as mediating mechanism for enhancing the technical capabilities of teams and, hence, increasing innovation performance. We propose to understand innovation as a learning process that is discovery-driven, enabled by empathy and fueled by failure. In this context, we highlight the role of individuals' cognitive and emotional powers as basic input for building a team's capacity for learning. Thus, we proposed that better understanding about the dynamics of collective action, shared cognition and emotional endurance for learning through feedback, evaluation and sense making would enable teams to better respond to the technical and procedural challenges when facing risky, ambiguous, and uncertain fast-paced environments. Finally, we suggest further research paths based on open questions, less explored paths, and conflicts and paradoxes
