Residential buildings are one of the key target sectors for energy and CO2 reduction. Research on the relationship between policy instruments and their effects is crucial for the continuous improvement of these tools. This thesis focuses on The Energy Performance of Buildings Directive (EPBD) as an EU flagship policy for technical improvements of the existing dwelling stock. At the core of the directive, an energy certificate is prescribed for all existing dwellings and is required to be present at moments of sale or rent. The certificate includes a theoretical consumption of each individual dwelling. However, theoretical models do not always correspond to reality with the desired accuracy. This large scale study explores the relation between the consumption assumed by the label certificate and the real consumption of the dwellings. It demonstrates that low performing dwellings have a theoretical consumption roughly two times the actual, while well performing dwellings consume one third less. These discrepancies are shown to have adverse effects on policy targets. Therefore, the thesis quantifies the causes of the discrepancies, looking into dwelling and household, as well as behavioural factors. There is a clear need for a more precise estimation of heating consumption on a broader, dwelling stock level in order to enhance the effectiveness of the current renovation policies. The thesis showed that using the current knowledge along with the growing amount of available data, there is enough motivation to reduce the performance gap by improving the predictions of actual dwelling consumption.