In the last years many studies have been carried out on the possible improvements of the in situ thermal conductance measurement; as well known, this has to be derived by recorded values of temperatures and heat fluxes. A big effort has been addressed in the implementation and comparison of different analysis methods. The aim of this paper is to widen the existent literature in the study of the influence of different kinds of input data on the final result. First the problem of analysing input data with significant drift in temperature is considered, adopting both nominal clean and noise affected data. Then the effect due to the presence of the Heat Flux Meter (HFM) on the thermal field of the testing element has been analysed, as well as the possibility of recording HFM surface temperature.These studies are based on Finite Element Method (FEM) simulations; both nominal clean and noise affected data have been considered as input. Finally, a difference in emissivity between the heat flux meter and internal plaster surface has been analysed by means of FEM simulations based on nominal clean data. An overall estimation of the occurred deviations in the different cases is shown. (C) 2011 Elsevier B.V. All rights reserved.