The topic of uncertainty quantification in particle image velocimetry (PIV) is recognized as very relevant in the experimental fluid mechanics community, especially when dealing with turbulent flows, where PIV plays a prime role as diagnostic tool. The issue is particularly important when PIV is used to assess the validity of results obtained with computational fluid dynamics (CFD). An approach for PIV data uncertainty quantification based on image matching has been introduced by Sciacchitano et al [1], where the contribution of individual particle images to the correlation peak is analyzed and the uncertainty is retrieved from the ensemble of particle image disparities. In this paper, the universality of the approach’s working principle is investigated via the application to a wide gamut of experimental data of flows ranging from laminar to turbulent regime and from subsonic to supersonic. Also a methodology for evaluating the performance of the image-matching approach in different experimental conditions is proposed.