1. Why would business process documentation be so important to the success of the HP project?

2. What other approaches could have been taken to the basic issues that HP was trying to resolve in this project?

3. Based on what you know of Dell, evaluate the company's best-of-breed strategy against a single-vendor strategy.

4. What are the risks of a best-of-breed strategy?

5. Which strategybest-of-breed or single-vendoris likely to remain valid for the long term?

REFERENCE FOR QUESTION 1 & 2

In 2002, HP launched a US$110 million ERP consolidation project and SAP rollout that it called GSO, named after the company's Global Supply Operations unit. The project was originally due to be completed within three years. By 2007, however, it was still not finished, although Peter Ginouves, director of finance at the GSO unit and a leader of the project, hinted that it should be done during that year. He added that "having three years to do something of this magnitude is extremely aggressive. Three years just zips by doing something of this scale." He maintained, however, that he still expected the project cost to remain within the US$110 million budget.

During 2003, HP suffered bad publicity after a poorly executed migration to SAP AG's ERP software in its server division. Its GSO project therefore followed hard on the heels of a major problem in the ERP implementation area.

The difficulties with the GSO project were blamed primarily on a lack of adequate internal processes. The GSO unit provides spare parts and repair services across the company and has annual revenue of US$2 billion. HP officials acknowledged that the aggressive schedule from the beginning didn't allow for adequate mapping of business processes or the implementation of change management capabilities.

Joshua Greenbaum, an analyst at Enterprise Applications Consulting in Berkeley, California, said, "The obvious object lesson is that the complexity of these projects requires even tech-savvy companies like HP to stop, slow down and make sure they are getting all the little details right." He added, "Redemption is possible if you catch the mistakes before you're too far down the road."

The goal of the GSO initiative is to consolidate 250 systemssome of them 20 years old and custom-written for HP, Compaq, Digital, and Tandem systemsaround a core ERP backbone based on SAP Enterprise 4.7. The existing systems also include varying instances of SAP, as well as an "alphabet soup of just about everything," Ginouves said.

Because of its difficulties with the previous project, the implementation team needed to establish immediate credibility. The 350-member team therefore scrambled to do a rapid deployment in 2002, installing applications such as SAP's materials management module but linking them to HP's legacy systems. In its haste, the team neglected to focus enough on business process management and failed to craft full end-to-end workflows, according to Ginouves.

By March 2004, it was clear that the project would take five years instead of three, and that it would likely only at best break even instead of achieving the expected 35 percent return on investment. HP developed a revised plan on the basis of achieving high double-digit returns on the remaining investment of US$40 million-US$45 million.

"Any large changes within organizations are challenging without a good change management program," said Dan Duryea, a supply-chain architect at the GSO.

Accordingly, the GSO decided to take a new approach that involved driving collaboration among business users and IT staffers to get a more thorough mapping of its business processes.

To achieve this objective, it adopted the NetProcess tool from IntelliCorp Inc. to map complete business processes, such as procuring a part and delivering it to a customer. The resulting process models can be summarized and shared with executives or segmented into task-level detail for use by HP's programmers.

Prior to adopting NetProcess, the project team had been using PowerPoint, Word, and Visio documentation tools, which didn't enable adequate communication among the employees implementing the system and, in any event, weren't designed specifically for business process documentation.

REFERENCE FOR QUESTION 3, 4, & 5

StrategyBest of Breed at Dell

When Dell Inc. decided to implement enterprise-wide systems, it chose the ERP software of Glovia International for inventory control, warehouse management, and materials management as part of its strategy. Dell also chose i2 Technologies' supply-chain management system and Oracle's order management system to implement what is known as a best-of-breed strategy. The concept of a best-of-breed strategy is to choose specific functionality from a variety of software packages instead of relying on one all-encompassing product. Then the packages need to be knit together by using middleware and other linkages.

The company had tried to implement SAP's R/3 ERP system in 1994 for manufacturing applications, but had felt it to be "too monolithic" to be able to keep up with Dell's changing business needs. After trying to implement the system over the following two years, it finally cut back the effort in 1996, using SAP only for human resources. Dell had tried to customize the product to match its business model, but found SAP couldn't keep up with the very rapid changes in this area.

Dell management pointed out that the Glovia software was scalable on Dell hardware, but also flexible enough to integrate with the other software they had chosen to run. In addition, they felt Glovia had the capability to break into smaller pieces for implementation. This was before SAP modularized its software to offer the same capability.

Dell undertook a global implementation of Glovia in 2000, with approximately 3000 Dell employees in the United States, Ireland, and Malaysia slated to use the software