Following the theoretical rationale for a role of policy evaluation in polycentric climate governance, what can be said about its actual role in light of theoretical expectations? The empirical data show that the theoretical expectations play out insofar as the elements that were hypothesized do exist in the context of the EU, Germany and in the EU, namely that both self-organized and non-self organized evaluation exist. The collective action dilemmas that feature at the very core of polycentric governance theorizing also materialize through evaluation, pointing to a significant role of the state, while non-state actors also contribute. Lesson-drawing may thus emerge as a function of evaluation, but the full theoretical potential of evaluation has not yet been realised, given the various deficiencies that both state-funded as well as society-funded evaluations contain.