This article proposes an empirical method to evaluate decentralised economic development programmes with heterogeneous characteristics implemented in different regions or states. The evaluation design developed in the article is a comparative analysis that operationalises differences in regional policy features and controls for exogenous factors that may affect outcomes independently from the programme intervention. The proposed method is illustrated and tested through the evaluation of the US Enterprise Zone programmes. The results of the analysis show that the proposed evaluation design is an effective tool for turning the heterogeneity of decentralised economic development programmes from a threat to the validity of the analysis into a great opportunity for testing the effectiveness of a variety of region- (or state-)specific policy implementation features.