The ever-increasing availability of information, together with the higher (time and financial) costs of data gathering, makes the use of pre-existing databases more and more convenient. The majority of the data gathered and recorded each day is not designed for research purposes however. It is still a task of each researcher to choose the relevant data in consideration of the research objectives, and to organize his own database according to his research purposes. The case study presented is the construction of a longitudinal dataset using synchronic data extracted from the administrative archive of the Sapienza University of Rome, and referred to the registered students’ careers. This dataset fits the purpose of studying the temporal dynamics and allows the analysis of specific phenomena (dropping-out, stopping-out, mobility, degree rates, etc.). Three different analysis on this dataset are presented, that highlight the usefulness of this kind of data structure: a quasi-experimental design, a sequence analysis and an event history analysis.