Purpose -The purpose of this paper is to evaluate the performance of 46 pharmacy retail stores and to analyze the impact of non-discretionary variables on the efficiency of the stores through dividing the decision making units (DMUs) into different categories. Design/methodology/approach -Basic CCR model is being used to evaluate the efficiency of each store. First, the DMUs are divided into three different categories namely severe, normal and advantageous depending upon the extent of competition that they are facing. Second, basic CCR, Phases I and II models are applied to evaluate efficiency, reference set and projections of DMUs within the categories. Lastly, all DMUs are evaluated together using the same models. Findings -For situations that are not under the direct control of the management, DEA with categorical DMUs is an appropriate approach for efficiency evaluation because through this approach one can do justice to each store which is a part of the study. Research limitations/implications -Inclusion of people factor such as customer satisfaction or employee satisfaction can lead to a better analysis and therefore, to a better practical approach. Originality/value -This study provides a framework for performance evaluation when both discretionary and non-discretionary variables are to be taken into consideration.