The estimation of the thermophysical characteristics of building elements based on in situ monitoring enables their performance to be assessed for quality assurance and successful decision making in policy making, building design, construction and refurbishment. Two physically-informed lumped thermal mass models, together with Bayesian statistical analysis of temperature and heat flow measurements, are presented to derive estimates of the thermophysical properties of a wall. The development of a two thermal mass, three thermal resistance model (2TM) enabled the thermal structure of the wall to be investigated and related to the known physical structure of two heavy-weight walls of different construction: a solid brick wall and an aerated clay, plaster, woodfibre insulation and gypsum fibreboard wall. The 2TM model produced good match to the measured heat flux at both interior and exterior surfaces for both walls, unlike a one thermal mass model (1TM); Bayesian model comparison strongly supported the 2TM over the 1TM model to accurately describe the observed data. Characterisation of the thermal structure and performance of building elements prior to decision making in interventions will support the development of tailored solutions to maximise thermal comfort and minimise energy use through insulation, heating and cooling strategies.