Retail industry has witnessed enormous growth in the past decade in developing countries like India, China, and Brazil, owing to the upswing in globalization, growing trends in e-commerce, multi-format retailing, and increasing penetration of the internet. The growth of opportunities, on the other hand, have intensified the competition. It is important for retailers to gain a competitive edge in the market through innovative strategies and continuous improvement. Meticulous planning and efficiency in operations are the drivers for economic sustainability and profitability of the business. Important prerequisites to gain efficiency and planning for improvement is the evaluation of base level performance, defining benchmarks and evaluating effectiveness of the efforts taken in this direction. The studies in this domain existing in the literature have analysed efficiency of retail stores as a black box transforming input to outputs. This approach lacks transparency and overlooks the subprocesses, their characteristics and internal interaction and can be addressed considering the transformation process in a two stage system. Our study addresses the issue and proposes a Bi-level Programming DEA approach to evaluate the relative efficiency of multiple retail stores considering a network structure operating in a Stackelberg relation and defining benchmarks for inefficient stores. The approach enables computation of efficiency of each sub-stage as well as the overall efficiency of the stores. The proposed approach is validated through a case study of Indian electronic retail chain.