Abstract: Learning processes carried out using ICT leave behind different types of identifying marks or “tracks” in computer systems in general and particularly in the log files of Web servers, the most commonly used system. Our starting assumption is that the study of these traces is essential in order to verify the efficacy and efficiency of these learning processes. In this paper we present the problems found in gathering the data required to analyse such learning processes, and also transforming this data for a multidimensional analysis. With this in mind we built a Learning Management System (LMS) to carry out the distance learning activities of the Faculty of Economics at the University of Trento, and adapted it to our research interests. We also present the multidimensional cube created by our team using the data collected from the LMS, as well as the techniques used for extracting and analysing the data.