This paper investigates the differences between the four modelling approaches using a school building as a case study. The four modelling approaches are GIS-based modelling, GIS-based modelling coupled with the input data used to produce Energy Performance Certificates (EPCs), template HVAC modelling in accordance with the National Calculation Methodology (NCM) in England and detailed component level HVAC modelling. The process of collecting and integrating modelling inputs from a range of databases and tools is demonstrated. The findings indicate that there are significant differences between the simulated annual energy consumption of the four baseline models and the measured data. Modelling approaches that incorporate higher levels of detail possess greater capability to simulate the actual operation of the system but tend to require more effort to calibrate in order to minimize additional uncertainty. Besides, the introduction of EPC input files may contribute to a rapid and reliable estimation of electricity consumption, however, the generation of EPC input files could have information biases introduced by the variability of modelers.