Spatially-explicit bottom-up energy models with detailed renewable energy representation are increasingly developed. In order to inform such models, we investigate spatial diffusion patterns of solar PV projects in 2'222 Swiss municipalities. Using a dataset of feed-in tariff and one-time subsidy recipients in 2016, we show that PV diffusion was spatially uneven throughout Switzerland in terms of four indicators: the number of PV projects per municipality, per 1'000 inhabitants, per unit of municipal electricity demand, and per unit of municipal land area. Urban-rural divide and exploitable solar PV potential are the key, but not the only predictors of the spatial heterogeneity in PV diffusion. The structure of the municipal economy, socio-demographic characteristics, regional spillover effects, and additional differences in local contexts, such as local policies, matter as well. Spatial diffusion patterns to some extent structurally differ across sub-national regions too, indicating that such empirical investigations are valuable in order to understand what can be generalized. We conclude with recommendations for developing and validating spatially-resolved energy models so that they capture realistic patterns of solar PV diffusion: gather, maintain and analyze spatial data on PV projects and develop robust modelling functions that do not only rely on PV potential. (C) 2019 Elsevier Ltd. All rights reserved.