© 2018 Elsevier Ltd Current building energy models are weak at representing the interactions between neighbourhoods of buildings in cities. The effect of a neighbourhood on the local microclimate is complex, varying from one building to another, meaning that neighbourhood effects on the airflow around a particular building. A failure to account for this may lead to the miss-calculation of heat transfer and energy demand. Current building energy simulation (BES) tools apply convective heat transfer coefficient (CHTC) correlations, which were developed by using a simplified model of wind flow that neglects neighbourhood effects. Computational Fluid Dynamics (CFD) techniques are able to model these neighbourhood effects and can be used to improve CHTC correlations. This work aims to develop a framework that couples CFD and BES tools to enhance the modelling of outdoor convective heat transfer in different urban neighbourhoods. A dynamic external coupling method was used to combine the benefits from both domains. Firstly, a microclimate CFD model was validated before the coupling stage using wind tunnel data. Secondly, the framework was tested using a benchmark model of a building block. Fully converged values of the surface temperature and CHTC were achieved at each time-step by the BES and CFD domains. The results highlight the importance of neighbourhood effect while the prediction of the hourly averaged external convection using coupling method can amend the simulation by up to 64% comparing to the standalone conventional BES models with DOE-2 CHTC approach.