Until recently, psoriasis was considered as a single disease entity. However, the discovery of major differences between early- or late-onset psoriasis suggests the presence of distinct disease phenotypes which may differ in their pathophysiology and in their treatment responsiveness. The objective of this study was to use exploratory data analysis methods to identify potential clinical psoriasis phenotypes without a priori hypotheses. A prospective questionnaire-based survey collected comprehensive informations on the main clinical characteristics of 1484 psoriatic patients. Six statistically different clusters of clinical symptoms were observed, corresponding at least to six different clinical psoriasis phenotypes. Moreover, discriminant functions allow patients to be assigned to one or other of these phenotypes. Our findings open the way to focus genetic, pharmaco-genetic, pathophysiological and therapeutic studies on more homogenous group of patients.