The nZEB objectives have raised the standard of building performance and changed the way in which buildings are designed and used. Although energy dynamic simulation tools are potentially the most suitable way for accurately evaluating and forecasting the thermal performance, they need several data inputs and user's knowledge that can affect the reliability of the results. It is precisely these two aspects that proved to be particularly critical, since the reliability of the ICT calculation tools has been widely proven in recent time.However, in order to foster credibility in sustainable architecture, bridging the gap between predicted and measured performance is pivotal to boost the building market towards energy efficiency and provide reliable data to inhabitant, investors and policy maker.The present research aims to identify and quantify the main factors that affect the energy performance gap through a detailed energy analysis carried out on a case study, which can be considered one of the first nearly zero energy residential complex built in Italy. Based on the analysis, the study identifies the main causes of the deviation between the calculated and measured data and demonstrates how it is possible to achieve very reliable models and, therefore, real buildings.Although the procedure traces a classic model calibration scheme, actually it consists of a verification of possible downstream errors mainly due to human factors, such as the provision of incorrect technical data or inappropriate operation.Some observations on the technical, management and regulatory gaps that may generate these errors are reported at the end of the study, together with practical suggestions that can provide effective solutions.