In Flanders, an obligatory software tool (EPR) is used to assess the energy performance of new buildings offering a simplified procedure to estimate the energy use for heating. This calculation approach is based on the principle of multiplying the building's heating demand with standardized (sub)system efficiencies. In this paper, the accuracy of this simplified approach is assessed for a traditional, hydronic heating system in non-residential buildings. To do so, integrated dynamic simulations are performed in TRNSYS for a series of building design variants with varying insulation quality, thermal capacity, window-to-wall ratio and orientation. From the integrated simulations, monthly subsystem efficiencies are deduced. Results show that the efficiencies are significantly influenced by the part load ratio. As however losses of efficiencies are noticed only in periods of low heat demands, the overall effect on the annual use is limited. Energy assessment by the simplified method is within an error of < 2.5 kWh/(m2.a) or < 10%. Therefore, the simplified approach as currently applied in the EPR calculation tool in Flanders is concluded to be suited for the calculation of the final energy use. An evaluation of tabulated values for the overall system efficiencies used in this simplified method is however recommended.