The identification of target firms and functional areas for strategic benchmarking

In recent years there has been an increased emphasis on quality, Total Quality Management (TQM),an d re-engineering. The practice of benchmarking is inherent to the success of any of these. Despite its obvious strategic implications, benchmarking has received little theoretical or analytical attention in the literature. Moreover, studies show that many firms are unsure about how to implement the benchmarking process. A critical aspect of benchmarking is the identification of who to benchmark against and what functional areas to benchmark. In this paper we show that Data Envelopment Analysis (DEA) can provide a structured methodology which, when used in conjunction with expert/managerial insight, can be a useful tool for providing necessary analytic support for the managerial practice of benchmarking. DEA uses multiple input and multiple output measures to assess efficiency. In this paper we first use DEA analysis to identify a subset of efficient firms which could be targeted for benchmarking. In a second stage, a constrained least absolute value (goal programming) method is used to ascertain input elasticities for the designated efficient firms in order to pinpoint important specific areas for benchmarking comparisons. We use macro-level data from the computer industry to illustrate this application of DEA and constrained least absolute value regression to benchmarking.