The variability in the clinical phenotype of Parkinson’s disease seems to suggest the existence of several subtypes of the disease. To test this hypothesis we performed a cluster analysis using data assessing both motor and non-motor symptoms in a large cohort of newly diagnosed untreated PD patients. We collected data on demographic, motor, and the whole complex of non-motor symptoms from 100 consecutive newly diagnosed untreated outpatients. Statistical cluster analysis allowed the identification of different subgroups, which have been subsequently explored. The data driven approach identified four distinct groups of patients, we have labeled: 1) Benign Pure Motor; 2) Benign mixed Motor-Non-Motor; 3) Non-Motor Dominant; and 4) Motor Dominant. Our results confirmed the existence of different subgroups of early PD patients. Cluster analysis revealed the presence of distinct subtypes of patients profiled according to the relevance of both motor and non-motor symptoms. Identification of such subtypes may have important implications for generating pathogenetic hypotheses and therapeutic strategies.