A growing literature in public policy, comparative politics and international relations has studied how the policies of one unit (e.g. country, federal state or city) are influenced by the policies of other units – that is, how policies diffuse. This article provides a meta-analysis of 114 studies, demonstrating persisting inconsistencies in the measurement of the mechanisms driving policy diffusion processes. Different indicators are used to measure the same mechanism, and the same indicators are used to measure different mechanisms. To improve this state of affairs, this article puts forward a conceptual structure that serves as a guide for the application of diffusion arguments, a starting point for theoretical refinement and a benchmark to assess measurement validity. In addition to paying more attention to the conceptual consistency of indicators, overcoming the problems currently found in the literature requires the construction of original, innovative research designs instead of the replication of widely used templates.