Spatial microsimulation models can be used for the analysis of complex systems. In this paper we make use of a spatial microsimulation model for the estimation of heat demand for Germany at a NUTS–3 level. The presented model creates a synthetic building stock by re-weighting the national microdata sample to small areas (NUTS–3) statistics with help of the GREGWT algorithm. Using the GREGWT method we benchmark the microdata sample to three different aggregation units (a) the building level (i.e. number of buildings); (b) families/dwelling units; and (c) individuals.