The Netherlands Cadastre, Land Registry and Mapping Agency (Kadaster) developed a nationwide solar potential and photovoltaic (PV) database to support the Dutch Climate Agreement targets. By integrating 3D geospatial modelling with GeoAI techniques, Kadaster assessed the rooftop solar potential for every building in the country. Two complementary approaches were adopted: (1) a 3D solar model, based on aerial imagery and elevation data, calculated solar radiation patterns multiple times per day and year; and (2) a deep learning–based object detection model automatically identified existing solar PV installations from high-resolution aerial images. The results from both methods were merged into a comprehensive geospatial database, enriched with building function and ownership details from official registers. This enabled policymakers, energy network operators, and planners to understand where solar energy can be deployed most effectively, track existing adoption rates, and plan for future infrastructure needs.