Abstract
We develop a firm‑level theory of collaboration architecture—how the breadth, geographic dispersion, and knowledge structure of partner portfolios jointly shape appropriable innovation and organizational scale. We theorize and find that knowledge coherence exhibits an inverted‑U relationship with patenting: moderate coherence supports integration and cumulative search, but overly coherent portfolios reduce knowledge variety. This concavity is steepest in hyper‑distributed architectures, where coordination and appropriability frictions are greatest. We further show that dispersion and diversity contribute differentially across the three architectures we identify—localized, distributed, and hyper‑distributed—and that enabling capabilities such as digitalization and research & technology organizations (RTO) connections condition the innovative value of broad or distant collaborations. Finally, we demonstrate that collaboration architecture shapes not only innovation output (patents) but also organizational scale (assets), clarifying when collaboration choices build inventive capacity versus when they support growth.
Please choose your submission type
Full Paper
Primary Track
Innovation