Shedding light on the As-Is vs. customized ERP debate :exploring the pragmatic practices of implementation-customization in a FMCG distribution company -

Abstract

As Enterprise Resource Planning (ERP) implementation continues to grow worldwide and in the Middle East, the scholarly work on ERP is dominated by ‘success factors’, ‘culture change’ or ‘best practice vs. local practice debate’. The latter stream of research has been concerned with should companies customize the parameters of ERP-specified processes to their own procedures or should they adopt them As-Is. Surprisingly, although this debate rages on in the scholarly literature, it is rarely informed by empirical research which looks at the details of practices of customization. Research that does so and involves Middle Eastern firms is even rarer. This research project takes on the latter gap in the literature and focuses on the study of customization practices up close in a Fast Moving Consumer Good (FMCG) firm in Lebanon in situ. The research focuses on the situated practices of customization during two critical stages of implementation life-cycle: Adoption (first experience with ERP) and Upgrade (maintenance of ERP after some time). These processes of implementation were analyzed based on the unfolding events and as experienced by key stakeholders to uncover key practices that were deemed influential in shaping the implementation outcomes. The findings of the research pointed to three key practices enacted in the implementation process which recurred both during the Adoption and Upgrade stages: decision making; test scenario development; and test script development. Furthermore, the findings also pointed out how in each of these practices there is variation in three factors, i.e., the amount of knowledge requirements, the degree of complexity, and the level of human resource (HR) involvement. Overall, Decision Making during Adoption Stage appeared to be more high-level concerned with financial considerations. Interestingly, in the Adoption Stage, both the Test Scenario Development as well as Test Script Development practices appeared to be more focused on generating somewhat