In this paper, we present a model-agnostic calibration methodology derived from IES' best practices and calibration guidelines. The calibration methodology relies on a 3-stage process that consist of (1) checking input priority matrix and SA results, (2) creating data-driven proles for high priority inputs, and (3) determining and deriving high-priority parameters. The process uses data analysis techniques, Sensitivity Analisys (SA) and optimisation tools to maximise model accuracy and minimise calibration efforts. IES headquarters, an oce building in the UK, is presented as case study. A model of this building used for ongoing commissioning has been calibrated at hourly level. Internal gains (i.e. lighting, equipment and occupancy) are derived from IoT sensors and included in the simulation as Free-From-Proles (FFPs). Room heating setpoints are included in the simulation as Parametric Proles (PPs). Sensitivity analysis and Optimisation-based parameter search is done by grouping spaces with similar-end use to minimise the number of parameters. Electricity and airtemperature calibrated to match utility data achieving a NMBE, CVRMSE and RMSE within recommended thresholds. Prediction error for air temperature outputs are minimised simultaneously to ensure that the model represents the building at space level. We show that use of metered data and automated tools can improve the quality of the model outputs at energy and space level with lower consultancy efforts.