Assessments of the applicability of equipment or the benefits of conservation measures within a buildingsector are frequent done using energy simulations of prototypical buildings with average physical and operational characteristics. Because of the large variations in size, hours of operations, energy-use intensity, and fuel-electric ratios even among buildings of the same type, vintage, and location, such an analysis risks the danger of missing "niche" markets for strategies such as cogeneration that may be attractive only under specialized conditions. The objective of this study is to analyze the impact that variations in building conditions have on a building's energy use patterns, and to develop procedures to account for these variations in assessing market potentials. The study utilizes a set of 481 prototypical commercial buildings developed for the Gas Research Institute (GRI) to study the applicability of cogeneration for commercial buildings in 20 U.S. urban areas. The physical characteristics, system configurations, and operating conditions of the prototypical buildings are based on either statistical data or engineering studies and chosen to reflect the average among the buildings represented by that prototype. In the current study, we first compare the adequacy of these "average" prototypes to capture the range of conditions within a building sector, and then purposely modify the building parameters of the prototypes to study how atypical conditions affect building energy use patterns. From this sensitivity analysis, we develop a procedure to account for variations in building parameters in assessing the market potential for specialized applications.