A data analytic benchmarking methodology for discovering common causal structures that describe context-diverse heterogeneous groups

Abstract

Modern organizations typically regard information and communication technologies (ICTs) as one of the significant direct or indirect inputs for achieving operational excellence and competitive advantage. Since the concept of competitive advantage involves a relative comparison of the performance of organizational entities, then the concepts of organizational capabilities, context, and benchmarking are relevant. In this paper we present a new multi-method methodology for benchmarking that explicitly takes into consideration context-specific factors impacting the performance of organizational entities. This novel methodology allows for obtaining actionable information, in the form of non-obvious common causal structures, for improving the performance of the less efficient entities vis-à-vis their more efficient counterparts. The new methodology is state-of-the-art and is novel because it explicitly takes into consideration the context within which the organizational entities perform. Such “context awareness” allows for expanding the universe of discourse within which the process improvement initiatives are usually considered, thus allowing to consider the impact of external to the process factors on internal to the process mechanisms. This methodology involves the creative integration of several Information Systems (IS)’ artifacts (i.e. multiple data mining methods) with Data Envelopment Analysis (DEA). We present an illustrative application of this methodology to an IS/ICT & Productivity research problem in the ‘developing’ countries context.