The most recent SIDIS measurements have challenged our understanding of factorization theorems in QCD. In order to perform reliable extractions of transverse momentum dependent functions (TMDs) it is essential to understand the sources of errors and weather they are under control. In this talk I will discuss about the issues encountered when comparing theory and experiment. I will also argue that careful examination of the features of the data is crucial for reliable TMD extraction. For concreteness, I will focused on the case of unpolarized SIDIS data.