Abstract Although intra-site spatial approaches are considered a key factor when interpreting archaeological assemblages, these are often based on descriptive, qualitative, and subjective observations. Currently, within the framework of research into spatial taphonomy and palimpsest dissection, several studies have begun to employ more quantitative and objective techniques, implementing tools such as geostatistics and geographic information system (GIS) methods. This is precisely the approach that the Abric Romaní team is following. In this work, we present GIS and geostatistics methods applied to the faunal and lithic assemblages from archaeolevel Ob, including an analysis of the spatial structure, the identification of clusters and sectors, size and fabric analyses, the projection of vertical profiles, and the reconstruction of a digital elevation model of the paleosurface. The results obtained indicate a clustered distribution, primarily concentrated into four dense accumulations. The predominance of remains < 3 cm in length and the absence of preferential orientations make it possible to rule out a generalised postdepositional movement affecting most of the site, although some local movement has been identified. The horizontal and vertical spatial analyses allow us to identify accumulations of a single material (lithic or faunal) in addition to mixed accumulations (lithic and faunal). Integrating all this data with the results of previous studies (zooarchaeological, refits, combustion structures, and partial lithic technological analyses), we evaluate and combine the interpretations proposed previously using different approaches, thereby improving the overall interpretation of the archaeolevel Ob. Finally, we also develop a preliminary comparison between Ob and some other levels at the same site (in particular M and P).