Analyses of data that include observations of multiple countries over longer periods of time—time-series cross-section data—have become an important tool for researchers in comparative politics, international relations, and many other fields. Such analyses are challenging, however. Researchers using time-series cross-section data may, for instance, experience that their results change radically once they make (seemingly) small changes to their regression models. In this Case Study, we explain how we went about our own analysis of time-series cross-section data related to welfare state reforms, why we chose our strategy, which problems we encountered, and how we dealt with them. We also provide a list of recommendations for researchers wishing to use time-series cross-section data that can help them make robust and valid inferences.