Nestl SAis a giant food and pharmaceuticals company that operates virtually all over the world. Headquartered in Vevey, Switzerland, the company had 2004 revenues of $76 billion and more than 253,000 employees at 500 facilities in 80 countries. Best known for its chocolate, coffee (it invented instant coffee), and milk products, Nestl sells thousands of other items, most of which are adapted to fit local markets and cultures.

Traditionally this huge firm has allowed each local organization to conduct business as it saw fit, taking into account the local conditions and business cultures. To support this decentralized strategy, it has had 80 different information technology units that run nearly 900 IBM AS/400 midrange computers, 15 mainframes, and 200 UNIX systems, enabling observers to describe its infrastructure as a veritable Tower of Babel. Interestingly, despite its size, the company has had no corporate computer center.

However, Nestl's management has found that allowing these local differences created inefficiencies and extra costs that could prevent the company from competing effectively in electronic commerce. The lack of standard business processes prevented Nestl from, for example, leveraging its worldwide buying power to obtain lower prices for its raw materials. Even though each factory uses the same global suppliers, each negotiated its own deals and prices.

Several years ago, Nestl embarked on a program to standardize and coordinate its information systems and business processes. The company initially installed SAP's R/3 enterprise resource planning (ERP) software to integrate material, distribution, and accounting applications in the United States, Europe, and Canada.

Nestl is working on extending its enterprise systems to all of its facilities to make its 500 facilities act as a single-minded e-business. Once this project is completed Nestl will able to use sales information from retailers on a global basis to measure the effectiveness of its promotional activities and reduce overstocking and spoilage caused by having products sit around too long on grocery shelves.

The experience of Nestl USA illustrates some of the challenges Nestl had to face in implementing enterprise systems. Nestl USA, an $8.1 billion subsidiary, in 2001 used to be organized as a series of brands, each operating independently. So, for example, the Stouffer's and Carnation units were separate companies, each owned by Nestl SA, the Swiss-based parent, but reporting to Nestl USA. In 1991, Nestl USA reorganized itself and the different brands were brought under the parent American control. However, the Nestl division headquarters were still dispersed, and each division was still free to make its own business decisions, although each one within the United States did report to Nestl headquarters in Glendale, California. The situation did not really begin to change until the spring of 1997 with the arrival of Jeri Dunn as vice president and CIO of the American company.

Dunn actually knew Nestl technology unusually well because of her long history with the company. In 1991, as associate director for application systems at Nestl-owned Stouffer's Hotels, she was sent to Switzerland to participate in an effort to establish a common worldwide methodology for Nestl projects. In 1995, she was promoted to assistant vice president of technology and standards for Nestl SA, and while there came to understand and agree with the value of establishing common systems throughout global Nestl because such a change would enable group buying which in turn would reduce costs. Dunn also realized that common systems would facilitate data sharing among subsidiaries. When she was moved to Nestl USA in 1997 at age 42, she found that her earlier recommendations from Vevey were mostly ignored. "My team could name the standards," Dunn said, "but the implementation rollout was at the whim of the businesses."

When she arrived, Dunn found that Joe Weller, the chairman and CEO of Nestl USA, wanted to integrate the company, although he was not an information technology specialist. Dunn joined with the executives in charge of finance, supply chain, distribution, and purchasing to form a team to study the company's strengths and weaknesses. They found many problems including the revelation that the company was paying 29 different prices for vanilla from the same vendor. Dunn's explanation was that "Every plant would buy vanilla from the vendor, and the vendor would just get whatever it thought it could get." She also realized that every division and every factory had assigned different names to the same product, so that the company could not even check on the situation. "We had no way of comparing," she said. When their studies were completed, they were given only two hours total to present their findings to Weller and the rest of the executives. Some of those reporting were upset with the time limit, and in the end they were given the whole day. Speaking later about the meeting, Dunn said, "[The executives] didn't know how ugly it was. We had nine different general ledgers and 28 points of customer entry. We had multiple purchasing systems. We had no clue how much volume we were doing with a particular vendor because every factory set up their own vendor masters and purchased on their own."

Soon after this meeting, the members of the team offered a three- to five-year plan for the necessary improvements. Central to the plan was the recommendation that the company install SAP, an ERP (enterprise resource planning) system. The team members expected the changeover to take three to five years. Dunn knew it was more than a software change, and she later said "We made it very clear that this would be a business process reorganization and that you couldn't do it without changing the way you did business." The long time period was the result of Dunn's expectation that "There was going to be pain involved, it was going to be a slow process, and this was not a software project." By October, Nestl had established a project team of 50 top business executives and 10 senior information systems professionals. They developed a set of best practices to become common work procedures for manufacturing, purchasing, accounting and sales. A smaller technical team was set up that took 18 months to examine all data for every item in all divisions and set up a common data structure for the whole company.

At first the project decided not to use SAP's supply chain software because that module was brand new and appeared to be risky. It turned instead to Manugistics for its supply chain module. The team did decide to use SAP's purchasing, financials, sales, and distribution modules. All of

these modules would be installed throughout every Nestl USA division. The plan was completed by March 1998 and development work began in July 1998. The project was called Best for "business excellence through systems technology."

In June 2000, Nestl SA followed the lead of Nestl USA and contracted with SAP to deploy purchase and deploy the new version of their software called mySAP.com. The new system will not only standardize and coordinate the whole company's information systems and business processes, but it also will extend SAP's enterprise software to the Web. The new system will allow each Nestl employee to start work from a personalized Web page linked to his or her job function. The employee's job is structured to conform to the "best practices" defined by SAP for 300 work roles. According to Jean Claud Dispaux, senior Nestl vice president for group information systems, "It is an exceptionally simple way to make sure that everyone does the same job in the same way." Nestl will also create up to five computer centers around the world to run mySAP.com enterprise financial, accounts payable, accounts receivable, planning, production management, supply chain management, and business intelligence software. Nestl publicity announced that the SAP contract would cost $200 million, plus Nestl would add an additional $80 million for installing all the system for the global company. However, a year after the announcement of the project, Anne Alexandre, an HSBC Securities analyst in London who covers Nestl, downgraded her Nestl recommendation. Her reason was her doubts about the success of the project. "It touches the corporate culture, which is decentralized, she said, "and tries to centralize it." She added, "That's risky. It's always a risk when you touch the corporate culture." Jeri Dunn agreed after her experience with Best.

The major problem that Best faced in the United States was that both Weller and most of the key stakeholders failed to realize how much the project would change their business processes. It soon was clear that they had created as many problems as they had solved. In fact a rebellion had already taken place when the team moved to install the Manugistics module.

The problem began during the early planning stage of the project when the staffs that would be directly affected by changes were not included in the key stakeholders' team. Dunn summed up the results, saying "We were always surprising [the heads of sales and the divisions] because we would bring something up to the executive steering committee that they weren't privy to." By the beginning of 2000 it was clear that nobody wanted to learn the new processes, that nobody wanted to make the changes. The lower level workers did not understand how to use the new system and also did not understand the changes. Nobody had been prepared for the new ways of doing things, and their only hope was to call the project help desk. Dunn said the help desk reached a phenomenal 300 calls a day. They did not want to learn the changes. Even the divisional executives were confused and angry. No one seemed willing to take the extra step to learn what to do. Turnover among the employees who were to use the Manugistics software to forecast product demand reached 77 percent. Those who remained found it easier to use their familiar spreadsheets.

In the rush to be done on time, the team had failed to integrate the various modules. Therefore, for example, while the purchasing departments used the appropriate systems and data names, their systems were not integrated with the financial, planning, and sales software. As a result,

when a salesperson gives a valued customer a special discount rate, it was entered in the new system, but the accounts receivable department would not know about it and would think the customer did not fully pay its bill.

The team finally called a halt to the project in June 2000. Nestl removed the project co-leader and Dunn took over as the sole project leader. In October, Dunn held a three-day offsite retreat with the key stakeholders and the business executives. It became clear that the deadline of January 1, 2000, had put too much pressure on the project, and as a result the members of the project team had lost sight of the bigger picture. They just focused on the technology. They now needed to integrate the existing components and to complete the work on the sales and distribution modules. Dunn also decided she now wanted to switch the supply chain module to the new SAP system because it had been improved enough since her rejection of it in 1998. By the time the retreat was ended, the team decided to start the whole project over again. It would first determine the business requirements and then decide on a new completion date, abandoning the earlier date. They also agreed to educate those affected so that all employees would know not only what changes were taking place but also why, how, and when those changes would happen.

The project team created a detailed design and project road map by April 2001. Nestl also assigned one person, Tom James, to be Best's director of process change, giving him complete responsibility for liaison between the divisions and the Best project. The team also began taking repeated surveys of the effect of the project on employees and how they were dealing with it. James and Dunn also began holding more meetings with the division heads. As a result of the information gathered in this way, James and Dunn determined the manufacturing users were not ready for the many changes, and so the rollout of that package was delayed for six months.

The new project appears to be paying off. All of Nestl USA are using the same software and data. The company not only has been able to produce better sales forecasts, but Dunn said the factories are following these better numbers. The company said it has already saved $325 million by spring 2002. And Nestl SA has learned from that project and expects to have an easier success with its project. And that, she says, is despite the fact that Dunn only had to deal with eight or nine autonomous divisions while the global headquarters was dealing with 80 autonomous countries to accomplish the same thing.

Nestle's global organization is benefiting from standardizing its data and business processes as well. By the end of 2004, about ten percent of Nestle's global food and beverage business was operating with standard processes, data and systems. Within the next few years most of the company's food and beverage business will undergo system implementations to bring them up to these standards.

Questions

Describe the problems and their impact on Nestl's operation. (4 marks)

What is the root cause of the problems? (3 marks)

If you had been in charge of the Nestl enterprise system project, what could you have done to prevent these problems? (8 marks)