Extant research on open innovation (OI) offers no systematic insight of how and why firms differ regarding the extent to which they conduct OI activities. Whereas past theoretical contributions have focused on explaining the externalisation of R&D activities as a result of firm-external factors, we focus on explaining this externalisation as a result of firm-internal weaknesses, specifically, impediments to innovation. Using the exploration–exploitation dichotomy as our theoretical framework, we develop hypotheses on how impediments to innovation influence the breadth and depth of OI. We then test these hypotheses by using an exceptionally large and detailed data set to estimate population-averaged panel models. Our results provide support for most of the hypothesised relationships. Further, they allow to identify four ‘archetypes’ of firms that differ significantly regarding the breadth and depth of OI and the importance of impediments. Finally, we discuss the significance of these findings for both academics and managers.