The Bayesian reweighting procedure is applied for the first time to a TMD distribution, the quark Sivers function extracted from SIDIS data. By exploiting the recent published single spin asymmetry data for the inclusive jet production in $p^uparrow p$ collisions from the STAR collaboration at RHIC, we show how such a procedure allows to incorporate the information contained in the new data set, without the need of re-fitting, and to explore a much wider $x$ region compared to SIDIS measurements. The reweighting method is also extended to the case of asymmetric errors, and the results show a significant improvement on the knowledge of the quark Sivers function.