Satellite observations from the Sentinel-5P TROPOMI instrument, combined with inverse modeling, provide a valuable resource for quantifying regional methane (CH₄) emissions. This study compares the emissions estimated from variational inversions in 2019 over Europe (0.5° resolution) assimilating three TROPOMI products of dry-column methane mole fractions (XCH₄). The SRON (v2.4, operational product), BLENDED (v1.0), and WFMD (v1.8) products are retrieved from distinct algorithms. They differ in coverage, error characterization, and XCH₄ spatial distribution. Results indicate that the largest contributions to XCH₄ differences may be attributed to aerosol scattering and sensitivity to albedo. The derived 2019 European CH₄ emission budgets show a relative increase of +2 % for SRON, and a decrease of -1 %, -33 % and -9 %, respectively, for BLENDED, WFMD and surface-based inversions. Seasonal emissions are highly correlated across the inversions. Spatial emission patterns and optimized boundary conditions are similar for the non-independent SRON and BLENDED but differ substantially from WFMD. Evaluation with independent surface stations shows error reduction for about half of the sites, with BLENDED performing best. However, no product is systematically closer to the emissions estimated when assimilating surface observations. Observing System Simulation Experiments (OSSEs) are used to disentangle the drivers of differences between the posterior emissions. They reveal that observation density and errors, but also averaging kernels and prior profiles play a key role in the inversion's capacity to constrain the emissions. Using consistent error definition and quality filters increases the consistency of the OSSEs, paving the way for more consistent emission estimates.