The paper presents a method to operationalise Jan Gehl's categorisation of dweller's activity patterns in public space using Foursquare data. The 'Urban Activity Wheel' method is instrumental in showing how location based social media data is beneficial to understand the distribution and variety of contemporary activity patterns. Re-organising both location-based social media data and statistical sources, unearths emerging activity patterns across scales from local to regional city making. Urban Activity Wheel shifts focus from the traditional functional analysis of urban space towards understanding activities and, thus, the human perspective of use, practices and new agencies. A specific analysis, the Shannon-Wiener Index of the complexity implemented on urban activities, gives further hints about the experiential qualities and development opportunities of urban spaces and neighbourhoods.