Relative Importance of Evaluation Criteria for Enterprise Systems: A Conjoint Study

Abstract

While a large body of research exists on the development and implementation of software, organizations are increasingly acquiring enterprise software packages (e.g., ERP systems) instead of custom developing their own software applications. To be competitive in the marketplace, software package development firms must manage the three-pronged tradeoff between cost, quality, functionality. Surprisingly, prior research has made little attempt to investigate the characteristics of packaged software that influence MIS managers' willingness to recommend purchase. As a result, both the criteria by which MIS managers evaluate prospective packaged systems and the attributes that lead to commercially competitive ERP software products are poorly understood. This paper examines this understudied issue through a conjoint study.

We focus on enterprise resource planning (ERP) systems, which are among the largest and most complex packaged systems that are purchased by organizations. In a conjoint study, 1,008 evaluation decisions based on hypothetical ERP software package profiles were completed by managers in 126 organizations. The study represents the first empirical investigation of the relative importance that managers ascribe to various factors that are believed to be important in evaluating packaged software.

The results provide important insights for both organizations that acquire such systems and those that develop them. The results show that functionality, reliability, cost, ease-of-use, and ease-of-customization are judged to be important criteria, while ease-of-implementation and vendor reputation were not found to be significant. Functionality and reliability were found to be the most heavily weighted factors. We conclude the paper with a detailed discussion of the results and their implications for software acquisition and development practice.

Keywords: Software selection, packaged software development, ERP systems, attributes, software design tradeoffs, and enterprise systems, conjoint, field study