The growing attention towards environmental sustainability in the pharmaceutical industry and increased awareness of the potential for improving energy performance are justified by the fact that the sector is energy intensive. However, the variety of the processes and the lack of data about production and energy consumption make it difficult to calculate Energy Performance Indicators (EnPIs) as much as to list Energy Performance Improvements Actions (EPIAs). This work elaborates data, such as final energy consumption and site characteristics, from 84 mandatory Italian Energy Audits (EAs) to calculate the mean and standard deviation of site-level EnPIs. Additionally, the suggested and implemented EPIAs are analyzed to describe achieved and potential savings. The results show what follows. In the typical pharmaceutical plant, around 70% of energy is used in auxiliary services, and its use is not related to production. For this reason, EnPIs calculated both with respect to mass production and plant surfaces have a mid-to-wide standard deviation; the mean primary energy EnPI calculated with respect to plant surface area is 0.38 ± 0.22 toe/m2. Most suggested EPIAs regard cold and hot energy production, as well as on-site energy production, from renewables and Combined Heat and Power (CHP) plants. The payback time is less than 4 years for many EPIAs, including both technical and managerial ones. According to the results, plant energy managers should calculate site EnPIs with respect to the site surface and increase monitoring of energy consumption at the process level. The last recommendation is also likely to be associated with more effective planning of EPIAs, allowing their introduction where the saving potential and economic indicators are more promising.