The dropout of university students is one of the topics of a broad interest in higher education institutions and government education departments. Recent changes in education methods, socio-economic conditions, and the growing limitations of face to face interactions make it necessary to have tools that allow us to consider a broad set of factors related to the dropout phenomenon. The objective of this article is to show the implementation results of a predictive information system (IS) for the prevention of university dropout in a higher education institution. The system allows the calculation of the risk of dropout per student and uses an alert generation procedure to coordinate interventions. The platform allowed measuring the impact of the intervention strategies on the permanence of the students. Likewise, it made possible the reorganization of the intervention process towards the students, prioritizing according to the risk level.