ZooMS has seen considerable development and increased used in the last decade. Most of the time, it has been applied by proteomic specialists, highlighting the potential of the method, but yet with little integration of the data produced in the general framework of zooarchaeological interpretations. The actual impact of ZooMS identifications on our understanding of past subsistence strategies, prey selection, transport decisions or butchering activities has yet to be fully uncovered. Taking a step back, we here frame theoretical considerations on the limits and benefits of ZooMS identifications, from the point of view of a “traditional” zooarchaeologist trained in the morphological identification of bone fragments in Paleolithic contexts, in charge of the analysis of large faunal assemblages. Issues such as bone fragmentation and differential identification (according to skeletal element and/or species) might have a dramatic impact on abundance measurements derived from ZooMS data. After highlighting the specificities of these biases, we explore ways of counterbalancing them using numbers of distinct elements, bone measurements and weights, and test our hypotheses using ZooMS data recently acquired on Middle and Early Upper Palaeolithic assemblages from southwestern France. Finally, through practical examples, we call attention to an additional key interest of ZooMS when it is applied to unidentified plotted bones: the definition of Post-Excavation Stratigraphies (cf. https://doi.org/10.21203/rs.3.rs-2405909/v1), and the increased reliability and resolution that ZooMS data can offer in our understanding of faunal changes when interpreted spatially.