Buildings represent 40% of total primary energy use in the U.S., and researchers have recognized the necessity of making office spaces more energy efficient. To reduce energy consumption, it is important to take into account buildings occupants' behavior into account. The difference between simulated, estimated and real power consumption in buildings reflects the need to search for more accurate ways to receive input from the users to better design energy saving policies. This paper presents then an energy saving process that, starting with a survey, classifies users by their job position. Based on the users' behavior, the process tunes the predefined profiles to achieve higher levels of savings without annoying them. We implemented part of the proposed process and statistically verified the existence of profiles, which in the future can be used to create personalized energy saving strategies and mechanisms.