Substantial theoretical, conceptual and empirical advances have been made in research on the implementation of EU policies during recent years. However, our findings have remained ambivalent and our theoretical insights disparate. It therefore seems high time to address some issues that cause concern in this type of research and to raise awareness of the limits of the various theoretical approaches and of the data commonly used. This relates to the challenges of operationalising and of choosing adequate indicators for the dependent variable (compliance). We also discuss promises and pitfalls of different types of data used in the field, such as official statistics on notifications and infringements published by the European Commission as well as mass surveys.