Commercial building flood losses significantly affect the Australian economy; however, there are not many models for commercial flood damage estimation and their results are not reliable. This study has attempted to derive and develop a new model (FLFAcs) for estimating the magnitude of direct damage on commercial structures. The FLFAcs – Flood Loss Function for Australian commercial structures, was calibrated using empirical data collected from the 2013 flood in Bundaberg, Australia, and considering the inherent uncertainty in the data sample. In addition, the newly derived model has been validated using a K-fold cross-validation procedure. The model performance has also been compared with the Flood Loss Estimation MOdel for the commercial sector (FLEMOcs) and the Federal Emergency Management Agency (FEMA) damage functions from overseas, as well as the ANUFLOOD damage model from Australia.The validation procedure shows very good results for FLFAcs performance (no bias and only five per cent mean absolute error). It also shows that ANUFLOOD, as Australia’s most prevalently used commercial loss estimation model, is still subject to very high uncertainty. Hence, there is an immediate need for a project to build new depth–damage functions for commercial and industrial properties.Awareness of these issues is important for strategic decision-making in flood risk reduction and it could amplify the cognition of decision-makers and insurance companies about flood risk assessment in Australia.