In the field of hardness measurements, a problem arises when trying to understand how different measurement parameters (speed of the indenter, force, thermal drift, etc.) affect the outcome of the measurement itself. Because the mathematical model defining hardness scales do not consider such factors, the simplest way to include additional influence parameters in the mathematical model is to introduce them linearly via sensitivity coefficients, which are obtained experimentally and thus characterized by uncertainties. Uncertainties of the sensitivity coefficients are in general not considered in the evaluation of the combined standard uncertainty of the hardness measurements. In this paper a general procedure is presented and applied to HRA and HRC measurements.