<p>The main approaches for investigating variability in stone artifacts are typological and technological. Technological approaches focused on understanding the entire knapping process dominate contemporary studies. One downside of standard technological approaches is the lack of standardization in reporting, leading to a lack of comparability. Moreover, multiple technological choices are frequently lumped under a single term, so that the underlying complexity is largely ignored and comparisons remain mostly descriptive and vague. For this reason, Gilbert Tostevin devised an approach that combines technological insights with statistical analyses, which makes it suitable for more formal comparisons between assemblages. In this study, we tried to improve several aspects of the approach, namely validity, sensitivity, interpretability of results, applicability, and replicability. We illustrate the utility of the approach by applying it to available data on the Middle to Upper Paleolithic transition. We argue that the revised version of Tostevin’s approach can be useful as a replicable method for exploring and explaining lithic variability.</p>