“reconstructing” Best Practices Embedded in Software Packages: an Actor-network Perspective

Management and consultants frequently preach the notion of “best practices” as a set of normative business practices that can be adopted to achieve superior organizational performance. However, the plausibility of a set of objective and context-free knowledge that can be readily acquired, stored, and transferable is questionable. Recent research has shown that “best practices” embedded in software packages can become potentially troublesome for local operations. The process by which organizations arrive at a consensus of what constitute best practices and how they recast the best practices embedded in software packages remains a black-box and presents an under-researched area of study. The research agenda is crucial, given the prevalent trend of package software adoption and that system vendors, as producers of such systems, are increasingly shaping the definition of industry best practices or standards for organizational consumers (Sawyer 2001). In this paper, we seek to open up this black box to describe and understand how the best practices embedded in software packages are reconstructed during the package implementation process. We adopt a narrative approach, to describe the negotiation and the modification of best practices during the implementation of a $14 million eprocurement system in a military organization across its 6,000 Army, Navy, and AirForce users. The adoption of a commercial-oriented software package by a large public sector organization presents a rich context to understand the process of appropriating best practices, especially given the explicit project policy of “no modification” despite obvious inherent incompatibilities. Specifically, the key best practices that were contested and renegotiated relate to the incompatibilities between the organization’s and software’s financial models (decentralized responsibility centers versus centralized vote management structure) and the acceptable security management practices (open/passive versus closed/active enforcement). Our longitudinal study uses the actant-network perspective to theorize the eventual eprocurement artifact as part of an ensemble of networked agencies and to analyze how best practices as socially enacted knowledge were inscribed in that artifact (Pentland 1995). More specifically, we trace the interactions among different agencies given their institutional histories, their conceptions of work problems and perceived goals of the technology, through their scripts of actions and rhetorical justifications used during negotiations, which eventually led to the eventual set of negotiated best practices that was inscribed upon the e-procurement artifact. By doing so, we highlight the politics and issues involved in localizing and appropriating purportedly “best practice” IT artifacts. As noted by Wagner et al. (2004), the process is not ruled by the majority – but the best practices are interpreted and constructed by a small group of actors who are able to materialize the preferred patterns of use. By opening up the black-box on the appropriation of best practices within package software, we hope to sensitize the IS