Buildings consume about 40% of the total energy in the modern world. As the demand for sustainable energy practices continues to rise, there is a growing need to enhance the efficiency of energy consumption in residential and commercial buildings. In this research paper, we introduce a framework based on data analysis that identifies and classifies energy inefficient buildings within a group. Bayesian inference is used to account for the randomness in daily energy consumption by estimating the distribution of factors that influence a building’s energy usage. The study analyzes real-time data from smart meters, such as electricity usage from various appliances, and compares them to other buildings with similar characteristics. Energy inefficiencies in a building can likely be traced back to their probable origins through fault detection and investigation. Such a report can be helpful in saving money, planning retrofit projects, reducing carbon footprint and is a step towards a sustainable environment. Our developed approach has been tested on a real world dataset and 33% of the buildings were found to be energy efficient. These results are impressive and demonstrate the effectiveness of the developed approach.