Previous literature has explored automatic personality mod-elling using smartphone data for its potential to personalise mobile ser-vices. Although passive modelling of personality removes the burden ofcompleting lengthy questionnaires, the fact that such models typicallyrequire a few weeks or months of personal data can negatively impactuser’s engagement. In this study, we explore the feasibility of reducingthe duration of data collection in the context of personality classifica-tion. We found that only one or two weekends can suffice for achievingstate-of-the-art accuracy between 66% and 71% for classifying the fivepersonality traits. These results provide lessons for practicing ”data min-imisation” – a key principle of privacy laws.