6: Case Studies – ERP and Information Systems

6
Case Studies

In this chapter, three case studies have been conducted: the first two case studies analyze the information systems (IS) of Hershey and FoxMeyer Drugs, both of which have already implemented enterprise resource planning (ERP) systems; while the third case study studies the strategy of Oracle as a software vendor. We briefly describe the IS evolution for each firm with an analysis of the impact of the research factors on relationships between ERP systems and IS integration or disintegration.

6.1 Hershey

Hershey, which was founded in 1894, sold several products: Kisses, Kit Kat, Reeses peanut butter cups, Twizzlers, etc. In 1996, Hershey chose to replace its legacy systems. There was growing demand from its retailers for suppliers such as Hershey to share their product delivery data so that the retailers could maintain optimum inventory levels and reduce costs.

Hershey selected SAP R/3, along with companion software from two vendors: Manugistics (supply chain software) and Siebel (customer relationship management (CRM)) = “TDEV–”. Software from systems, applications and products for data processing (SAP) included modules for finance, purchasing, materials management, warehousing, order processing and billing. Manugistics provided software for transport management, production, forecasting and scheduling. Siebel was to support Hershey in managing customer relations and in tracking the effectiveness of the company’s marketing through a pricing promotions module. Hershey had used software from Manugistics earlier in its mainframes = “evolution strategy of existing systems (ESES–)”.

At this period, even if Hershey would have preferred purchasing SAP’s supply chain management (SCM) and CRM modules over those of Manugistics and Siebel, it would not have been possible, because SAP did not offer these modules to clients in 1996. In fact, before 2002, the vendors’ strategies (SAP, Oracle, etc.) did not take into consideration the new modules that seem vital today (CRM, SCM, etc.). At that time, these modules were commercialized by certain Best of Breed companies such as Siebel and Manugistics. SAP was keeping its ERP 1st G without any expansion toward a 2nd G, and we therefore define the evolution strategy of this ERP vendor (ESEV) information technology (IT).

IBM Global Services was chosen as the integrator. Accenture and SAP helped the IT teams to complete implementation. In order to speed up the project (30 months instead of 4 years), Hershey then decided on a Big Bang approach, where several modules were implemented simultaneously, instead of a phased approach designed to find and correct bugs before moving on to the next phase. The employees needed to be trained on the system functionality and also on how the different modules interacted with each other. They were overloaded, as they had to learn the intricacies of not one, but three new systems, and bring out the required revisions in their activities = “PMER–”.

Hershey’s former CEO and Chairman Kenneth L. Wolfe told Wall Street analysts during a conference call at the end of 1999 that the company was having problems with its new order-taking and distribution system. Slowly, problems pertaining to order fulfillment processing and shipping started to arise. Several consignments were shipped behind schedule, and even among those, several deliveries were incomplete. Customers began switching to products of competitors such as Nestlé and Mars. Moreover, orders from many retailers and distributors could not be fulfilled, even though Hershey had the finished product stocked in its warehouses. The company said that the new systems were not transmitting order details to the warehouses and the data flow between different applications needed to be corrected in order to fix the problems = “INTE–”.

“The reduction in shipments resulted primarily from difficulties in order fulfillment (customer service, warehousing, and shipping) encountered since the start-up of a new integrated information system (IIS) and new business processes during the third quarter of 1999” (Hershey, annual report 1999). The results of failed ERP system implementations were immediate, with a significant drop in the revenues for third quarter of 1999. Annual revenues for 1999 were US$150 million less compared to those in 1998, a drop of 12% = “ECCO–”.

Experts and analysts were of the view that:

  1. – with three different vendors working on the system, it would have been better if Hershey had chosen to roll out each system successively and then check the integration issues =“INTE–”. “There were three cooks in that kitchen. That’s why there is so much finger-pointing going on” [OST 00];
  2. – problems could have been avoided if there had been more focus on training = “PMER–”;
  3. – “these systems tie together in very intricate ways, and things that work fine in testing can turn out to be a disaster” (Shepard, Vice president for AMR research 1999). Aside from the fact that SAP itself was complex to install and run, Hershey attempted to implement two other applications along with it, making the whole exercise still more complex = “COER–”;
  4. – with more than one package in operation, there was a sharp increase in the number of touch points and interfaces, and implementation-related problems were almost inevitable. There were problems integrating SAP with Siebel and Manugistics =“INTE–”.

The upgrade of the failed project began in July 2001 and was finished in May 2002. Hershey redesigned the process and began working with SAP 4.6. According to Joe Zakutney, Director, SAP at Hershey, “the company’s success in its second round of ERP implementation was attributable to strong program management and executive leadership diligent planning and an extensive testing and training plan [WEI 02]”= “PMER+”. Within 11 months, the system was implemented successfully.

The lack of interoperability within the framework of the IS was one of the primary reasons for the original project’s failure. In order to address this issue, Hershey improved the interface between distribution function and SAP = “INTE+”. Hershey said it was able to make more than 30 improvements to its core business processes. The company cited enhancements such as the automation of pick-list processing and materials management invoice verification, plus credit processing for distributors to military customers. These improvements have helped reduce costs = “ECCO+” and speed up processing times.

6.1.1. Case outcome and evaluation

Hershey is an interesting case study for our research, because, after a first round of an unsuccessful ERP system implementation (a kind of IS disintegration), a second round enabled Hershey to achieve a successful ERP system implementation (integration).

Before the successful updating in 2001, the integration problems between the three subsystems (SAP, Siebel and Manugistics) were related to the unreliability of the interoperability and to the ERP system’s complexity. Therefore, “INTE–” and “COER–” have been given as values. In fact, prior to 2000, before the successful update, the interoperability was not sufficiently developed by the ERP system vendors and to this day, they are still working to improve it. Regarding the complexity, we consider that there is still much progress to be made by the ERP system vendors. However, after many years of usage and experience, SAP, Siebel and Manugistics became less complex for Hershey’s employees.

Although improvements have been made, even after the successful updating in 2001, we cannot give the two factors “INTE and COER” positive values. In fact, we know that in September 2005, Oracle Corporation acquired Siebel. As we explained previously (see factor ESEV), the redesigning of Siebel into Oracle could be performed only in a more or less integrated way, which is complex enough. Therefore, the interfacing of Oracle with the rest of the architecture (mainly SAP) is not easy to implement. This is why we give values between “– and +” for INTE and COER.

Regarding the evolution strategy of ERP vendors, we already said that in September 2005, Oracle Corporation acquired Siebel, which thus went out of business = “ESEV–”. However, by instituting an external acquisition, Oracle allows us also to give this factor the value of “ESEV+”. Hershey’s update into new versions, with all of its subsystems within a large vision, became even more difficult because of the expansion strategies of their ERP system vendors. As a result, we also give values between “– and +” for ESEV.

All of the research factors that have been discussed and detected by the literature review have been confirmed by the Hershey case study. Although the firm’s aim was to implement an IIS, this case study illustrated two types of IS evolution (first a disintegration and then an integration):

  1. – the first round of this project management (before 2000), which was a failure, did not allow this goal to be reached. The negative values, which have been detected for all factors, have led to a disintegrated information system (DIS) instead of the desired IIS. We can conclude that the combination of negative values of all factors caused the ERP system’s evolution to promote IS disintegration;
  2. – the second round of the project (after 2001), which was a success, allowed the architecture to transform from a DIS to an hybrid integration of IS (HIIS). The values of certain factors have changed (economic crisis and competitiveness (ECCO), project management ERP (PMER), complexity of ERP (COER), interoperability of the ERP (INTE) and ESEV)), while other values have not changed (total dependency on the ERP vendor (TDEV) and ESES). We can conclude that the new combination of values favors an HIIS.

6.2. FoxMeyer Drugs

FoxMeyer Drugs was a $5 billion company and the fourth largest distributor of pharmaceuticals in the United States before the fiasco. With the goal of using technology to increase efficiency, the Delta III project began in 1993. FoxMeyer purchased SAP in December 1993, as well as warehouse-automation from a vendor called Pinnacle = “TDEV–” and chose Andersen Consulting to integrate and implement the two systems. Implementation of the project took place during 1994 and 1995. The firm was driven to bankruptcy in 1996 = “ECCO–”, and the trustee of FoxMeyer announced in 1998 that he was suing SAP as well as Andersen Consulting for $500 million each [CAL 98]. The reasons for the failure were the following:

  1. – the processes of the business practices were not reengineered to be adapted to fit the new design integration (SAP and Pinnacle). Some users were not fully committed since the project threatened their jobs. This meant resistance to change and rejection of the tool by users. In addition, a shortage of skilled and knowledgeable personnel was crucial. There were over 50 consultants at FoxMeyer, many inexperienced, and the turnover was high. According to FoxMeyer, Andersen used the project as a training ground for “consultants who were very inexperienced” [COM 98]. Besides, despite warnings from Woltz Consulting during the early stages of the project that an 18-month schedule for the entire implementation to be completed was unrealistic = “PMER–”, Delta project went ahead [JES 97];
  2. – the decision to go with two different vendors = “TDEV–” for two of the company’s most important business systems was “an error in information processing” [KEI 95]. This added still greater complexity to an already challenging situation [JES 97], given the integration’s original complexity (SAP and Pinnacle subsystems). The warehouse automation multiplied the project risk, and interactions between SAP and Pinnacle’s automation took FoxMeyer into uncharted waters = “INTE–”. Using just one vendor in the first phase would have reduced the risks and complexity of the project = “COER–”.

6.2.1. Case outcome and evaluation

We think that the subsystem of warehouse-automation was not bought from SAP for one of the two reasons:

  1. – this module was not taken into account within the evolution strategy of SAP in 1994;
  2. – FoxMeyer did not want to be dependent on one ERP vendor.

This is why we gave both factors the negative values = “ESEV– and TDEV–”.

6.3. Oracle Corporation’s E-Business Suite

This case study is prepared from many interviews that were retrieved from the Internet of certain directors of Oracle: Larry Ellison, a co-founder of Oracle Corporation and Jimmy Anidjar, who was Senior Vice President Oracle France and MEA (Middle East and Africa). Some responses of these interviews are given in the study. Therefore, this case study represents the vendor’s point of view combined with our analysis.

Oracle is a multinational computer technology corporation based in the United States that specializes in developing and marketing computer hardware systems, databases and ERP systems. “In 2004, we bought PeopleSoft. We had three ERPs, two CRMs, two HR systems and it became even more complicated with the arrival of Siebel. Since the beginning of 2005, Oracle has made no less than 27 acquisitions. The latest is that of Hyperion in 2007 which was a Best of Breed in Business Intelligence (BI). We knew we had to create a new generation of ERP, assembling the best features of PeopleSoft, JD Edwards, Oracle E-Business Suite, Siebel, decision making and advanced collaborative tools = “ESEV+”. Our goal is to help customers replace application silos by an ERP wide scope announced in Q1 2011” [ELL 10]. Oracle sought to standardize its offerings and build a growing number of products on the same foundation.

“It took five years and a lot of work to get there. The Suite contains 10,000 unique integrated processes and the final version will include 100 different products. We had never done this before, and I hope we will not have to do it again = “COER–”” [ELL 10]. “How then can one design a new ERP based on many ERPs (different technologies and formats)? After heavy rewriting and production of a new version of the ERP which takes into account, in a more or less integrated way, the modules recently acquired, we obtain a wider perimeter of the ERP. But for firms, updated versions become very cumbersome as the new redesigned ERP is technically quite different from the previous and such a scenario could happen again in the future” [PRA 08].

Oracle also seeks to promote the interoperability of the final version, which contains an application integration architecture that includes middleware interoperability. This orchestration allows the linking of functions either from the existing software package or other solutions that may or may not be Oracle software. “We know that the introduction of new modules, especially after the acquisition of the vendor that made the software, by a competitor generates a lot of difficulties without guaranteed result” [DES 04]. Even though the interoperability’s improvement was taken into account by Oracle, we cannot know if this interoperability was completely reliable. For this reason, we choose to give values between “– and +” for INTE.

Oracle’s clients can decide whether to purchase the whole package or only some modules = “TDEV–”, but whatever the migration tools provided by Oracle, we can expect that migration is an important project = “PMER+”, as the ERP is built on a new foundation. New customers are oriented to this new coherent offer (if new clients would like to buy the whole package, a total overhaul will be necessary to replace their existing systems = “ESES+”). Old customers can keep old products, which continue to evolve (if old clients would like to buy only a part of the package, an urbanization would be sufficient = “ESES–”). We note that the factor ESES adopted by firms is taken into account by the vendor.

6.3.1. Case outcome and evaluation

This case study permits an analysis of the evolution strategy of an ERP system vendor. We think that the implementation of Oracle, which represents an ERP 2nd G developed by the vendor according to a strategy of external acquisition, promotes an HIIS rather than a total integration of IS (TIIS) (see Table 1.1).

6.4. Summary of case studies

Contrary to a simple software or subsystem, a complicated ERP system needs a considerable amount of training. The tests, especially integration tests, have not been carried out correctly. Therefore, a “PMER–” has generated an “INTE–”. It seems, in the late 1990s and early 2000s, the enlargement of the ERP perimeter was limited to an ERP 1st G without any new modules such as CRM, SCM, BI, etc. As a result, the only possible value, at this time, for the ESEV is “–”. Contrary to the main goal for the two firms that targeted an IIS implementation, some combinations of factor values have led to a kind of IS disintegration. Other combinations for Hershey after their successful upgrade have allowed a kind of integration by an evolution from a DIS to an HIIS.

The case studies generally confirm the reliability of research factors explored previously, as well as the suggested correlations between these factors. These case studies also allow us to observe relationships between the research factors and IS integration or disintegration. According to a given combination of the mentioned factors’ values, different architectures are observed (especially DIS and HIIS). Table 6.1 resumes the principal ideas of these three cases.

Table 6.1. Factors affecting the relationships between the evolution of ERP systems and IS integration or disintegration according to the case studies

TDEV ECCO COER ESES ESEV PMER INTE IS
Hershey before success DIS
Hershey after success + Between – and + Between – and + + Between – and + HIIS
FoxMeyer NA DIS
Oracle Corporation NA Between – and + + + Between – and + HIIS