The growing availability of mobile devices produces an enormous quantity of personal tracks which calls for advanced analysis methods capable of extracting knowledge out of massive trajectories datasets. In this paper we present an experiment on a real world scenario that demonstrates the strong analytical power of massive, raw trajectory data made available as a by-product of telecom services, in unveiling the complexity of urban mobility. The experiment has been made possible by the GeoPKDD system, an integrated platform for complex analysis of mobility data. The system combines spatio-temporal querying capabilities with data mining and semantic technologies, thus providing a full support for the Mobility Knowledge Discovery process.