This study develops a set of prediction models to estimate energy usage for commercial blocks in urban areas. Commercial blocks in Taipei City, Taichung City, and Tainan City of Taiwan were selected as research subjects, and a total of 93 blocks were surveyed. With the survey of land use type and intensity, building usage, and actual building energy usage of the blocks, a set of commercial block energy usage prediction models were proposed and tested, and two commonly used land use variables (the floor area ratio and total floor areas) were included as independent variables in these models. The average number of floors and an indicator of commercial activity intensity were then employed to help define the level of energy usage intensity for the blocks surveyed in order to refine the models, and the R2 for both energy usage prediction models increases to 0.94 and 0.92, respectively. This study shows that land use variables can be utilized to rapidly estimate energy usage of commercial blocks in urban areas, and the finding should be useful for decision-makers to formulate policies on urban energy management and land use planning.