Judy E. Scott, The University of Texas at Austin, Judy.Scott@bus.utexas.edu Abstract This interpretive case study of FoxMeyer Drugs' ERP implementation is based on empirical frameworks and models of software project risks and project escalation. Implications of the study offer suggestions on how to avoid ERP failure. warehouses, the transition to the first automated warehouse was a disaster. Disgruntled workers damaged inventory, and orders were not filled, and mistakes occurred as the new system struggled with the volume of transactions. $34 million worth of inventory were lost (Jesitus 1997). Second, the scope of the project was risky. FoxMeyer was an early adopter of SAP R/3. After the project began, FoxMeyer signed a large contract to supply University HealthSystem Consortium (UHC). This event exacerbated the need for an unprecedented volume of R/3 transactions. Although, prior to the contract, testing seemed to indicate that R/3 on HP9000 servers would be able to cope with the volume of transactions, in 1994 R/3 could process only 10,000 customer orders per night, compared with 420,000 under FoxMeyer's original mainframe system (Jesitus 1997). Third, the execution of the project was an issue due to the shortage of skilled and knowledgeable personnel. FoxMeyer did not have the necessary skills in-house and was relying on Andersen Consulting to implement R/3 and integrate the ERP with an automated warehouse system from Pinnacle. Although at the height of the project there were over 50 consultants at FoxMeyer, many of them were inexperienced and turnover was high (Computergram International 1998). Finally, the environment quadrant of the risk framework includes issues over which project management has little or no control (Keil, Cule, Lyytinen and Schmidt 1998). Although FoxMeyer must have realized the project was in trouble, its perceived dependence on consultants and vendors prevented it from seeing how it could gain control. Since FoxMeyer was competing on price, it needed a high volume of transactions to be profitable. Yet with the UHC contract "the focus of the project dramatically changed", contributing to rising project costs (eventually over $100 million), lowering FoxMeyer's already narrow margins and erasing its profitability. Given the high level of risk, why did FoxMeyer initiate the project? Furthermore, why was the project allowed to escalate to the extent of contributing to FoxMeyer's bankruptcy?
FoxMeyer Drugs was a $5 billion company and the nation's fourth largest distributor of pharmaceuticals before the fiasco. With the goal of using technology to increase efficiency, the Delta III project began in 1993. FoxMeyer conducted market research and product evaluation and purchased SAP R/3 in December of that year. FoxMeyer also purchased warehouse-automation from a vendor called Pinnacle, and chose Andersen Consulting to integrate and implement the two systems. Implementation of the Delta III project took place during 1994 and 1995. According to Christopher Cole, chief operating officer at Pinnacle, the FoxMeyer mess was "not a failure of automation. It was not a failure of commercial software per se. It was a management failure" (Jesitus, 1997). Perhaps management had unrealistic expectations. Did management expect technology to be a "magic bullet"? (Markus and Benjamin 1997a, 1997b). In reality, it was the opposite. FoxMeyer was driven to bankruptcy in 1996, and the trustee of FoxMeyer announced in 1998 that he is suing SAP, the ERP vendor, as well as Andersen Consulting, its SAP integrator, for $500 million each (Caldwell 1998, Stein 1998).
The Delta III project at FoxMeyer Drugs was at risk for several reasons. Using a framework developed for identifying software project risks (Keil, Cule, Lyytinen and Schmidt 1998), this study classifies the project risks at FoxMeyer into (1) customer mandate, (2) scope and...