The performance gap is a measure of the difference between design assumptions and actual in field data. Estimating terms, specifically the operational hours of use, when making such design assumptions in order to predict the impact of lighting upgrades can potentially result in either an overestimate or underestimate of the savings to be made. In this paper, the background of performance gap measurement is outlined and field measurements are gathered and applied retrospectively to lighting upgrades in classrooms. The lighting upgrade projects in three university buildings and their assumptions are explained in relation to the operational hours proposed using the industry ‘Energy assessment and reporting method’. We then describe a simple and relatively inexpensive means of taking in field measurements using small unobtrusive environmental loggers to record the lighting use and occupancy. This method, which can be implemented prior to upgrade works or energy efficiency retrofits, reveals different patterns of lights on use, occupancy and booking hours from those assumed by a priori estimates. These results also suggest a valuable distinction between lighting on hours and occupancy hours not captured in the current ‘Energy assessment and reporting method’. We conclude that the estimates the industry uses in calculating energy efficiency upgrades should be accompanied by clear and adequate information about occupancy use. In our study, the consequence of reporting energy savings using assumptions and estimates in calculations for classrooms resulted in limited overall differences in the savings achieved in practice.