Current debates around the 'performance gap' have highlighted the need to study building occupancy patterns to improve design solutions and better understand space utilisation. However, capturing occupancy data is resource intensive. There is a need for solutions that gather real-time occupancy data while maintaining the users' privacy. In response to this challenge, this paper discusses applying a thermal imaging-based method for measuring occupancy in buildings and generating behavioural insights. A longitudinal analysis of the occupancy patterns over a full academic year is conducted for a university library building in the UK. The granular data collected through the thermal imaging analysis reveal insights into the building's occupancy patterns over academic terms and vacation periods. The findings debunk conventional conceptions of library use during weekends/weekdays and terms/vacations. The application of thermal imaging sensors to monitor occupancy within the library building suggests the potential use of real-time data to improve the library's space and organisational management. The paper makes a case for having an occupancy monitoring strategy in place that corresponds to the data needed for making effective interventions.