The purpose of this research is to propose an internal comparison approach to inform energy efficiency of higher education buildings. Traditionally, the median values of Energy Use Intensity (EUI, unit: GJ/m2) from the Energy Star program can be used as a benchmark to assess buildings of the same category and primary function. However, higher educations buildings tend to have mixed usages (e.g., lab, classroom and office). Thus, instead of labelling a building with a single primary function, this work examines the percentages of floor areas of different usages for each building (e.g., 25% classroom and 30% office). Then, the data of floor-area percentages are analyzed using linear regression and hierarchical clustering. Based on our case of 24 campus buildings, we have classified four types of floor areas: lab, public services, school services and other. The regression model can correlate the EUI with these floor-area types with R2 = 89.62%. Based on both regression and clustering results, we employ the analyses of residuals and building groups to investigate the energy efficiency of our building stock. This study has shown that the proposed method can provide some insights for facility management to investigate and prioritize the energy issues of the building stock. Unique insights include the energy efficiency of lab-intensive buildings and the identification of inefficient buildings which are less obvious in the original EUI comparison.