The existing building patrimony is responsible for 36% of the global energy use and 37% of the greenhouse gas emissions. It is hence a major challenge to improve its energy performance. According to the Renovation Wave, the average annual renovation rate should be doubled by 2030 up to 3% and deep energy renovations should be encouraged. The Belgian city of Leuven works towards this target and is even more ambitious, setting their goal on becoming climate neutral by 2050. The strategy investigated in this study is to increase the renovation rate by clustering renovations, which is challenging since the Belgian building stock is highly privatised. Based on a thorough literature study, this paper examines various methodologies for building stock modelling. The main focus is comparing the required input data with the data availability, handling the data gaps, and defining their influence on the model’s accuracy. The findings are applied to Leuven by analysing the main drivers to cluster renovation measures. However, many data gaps appeared, leading to the selection of a GIS-enhanced archetype model enriched by energy data as the most suitable approach. To avoid misinterpretation due to differences in data quality, transparent reporting in stock modelling is recommended.