Introduction

In the 1990s many firms turned to software packages when they replaced the older, often home-built systems in their application portfolios. They chose to buy, rather than build, their new systems. Broadly, they followed the prevailing wisdom to downsize themselves so as to focus on their ‘core competencies’ and to outsource other tasks to the market. Many found no compelling logic for continued in-house development of their application software. Some, with their staffs already reorganized and reduced through reengineering, were simply no longer in a position to undertake major development tasks.

Replacement of systems had in the 1990s become a priority for many firms. These organizations felt increasingly burdened by their ‘legacy systems’. Their support staff were often committed to a maintenance task for which the costs were all too apparent and the benefits appeared to be marginal and remedial. Their home-built software also often resided on expensive mainframes widely viewed as heading rapidly toward obsolescence. Further, users found this software increasingly cumbersome to work with in comparison to their newer PC-based tools with their graphical user interfaces.

It was in this context that the 1990s gave rise to ERP (enterprise resource planning), a significant innovation in the packaged software market. The vision for ERP was first articulated by the Gartner Group (Wylie, 1990). It took an ‘enterprise-wide view’ of traditional application software, allowing for internal integration of the technical and business core, as well as external integration with business customers.

Introduction

An increasing number of multinational enterprises (MNE) have adopted ERP systems in the hope of increasing productivity and efficiency as a means of leveraging organizational competitiveness (Davenport, 1998; Wagle, 1998). While some are starting to harvest the benefits from their initial investments, others are still struggling to release the promised potential of their ERP systems. This can be seen as an illustration of the ‘productivity paradox’ (Fitzgerald, 1998), that is that firms face a significant problem in measuring the return on their IT investments. While there have been several accounts that have examined the adoption of technology, including ERP, few of these accounts have considered this from the perspective of cross-functional knowledge integration (Grant, 1996). Against this backdrop, a case study was conducted as a means of exploring and theorizing the dynamics of knowledge integration underlying the process of ERP implementation. In this paper we focus not only on presenting the theoretical framework but also on describing the stages that were undergone to derive this framework.

Current Debates and Perspectives

There is a great deal of literature conceptualizing the phenomenon of how firms implement new IT systems. This study draws upon research in four distinctive areas, including (1) the development of technology, (2) the process of technology implementation, (3) enabling and inhibiting factors, and (4) management of process innovation.

Organizations invest in enterprise system software from vendors such as SAP, Oracle, PeopleSoft, Siebel, and i2 Corporation to gain access to powerful computer-based information systems more cheaply than through custom-built software development. What they acquire is a highly flexible software product, containing solutions to the needs of many of the vendor's existing customers, that impounds deep and detailed knowledge of many good and less good ways of conducting a wide range processes in a broad range of industries. The difficulty for the licensing organization is to identify the right combination of processes for its own changing needs and to implement those processes in its own organization.

Enterprise systems and the internet were probably the two most important information technologies to emerge into widespread use in the 1990s. According to Technology Evaluation.com, the total revenue from the enterprise system software and services market was US$18.3 billion in 1999 and US$19.9 billion in 2000 (Gilbert, 2000; Jakovljevic, 2001). Enterprise system implementation costs are often reported to be five to ten times the cost of software licenses (Davenport, 2000; Scheer and Habermann, 2000). If so, organizations worldwide spent something like US$100 billion per annum on enterprise systems in both 1999 and 2000. In short, organizations around the world have made huge investments in enterprise systems in the past decade.

Introduction

Recognizing the need to present a single face to global customers, to respond rapidly to customer demands, and to seek out economies of scale, business executives have been for some time regularly examining the capabilities offered by information technology (IT). As a result, from the mid 1990s on, many firms have been replacing the legacy systems that form the base of their information processing capabilities with enterprise resource planning systems (ERPs). ERPs replace a firm's disparate transaction processing systems with a single, integrated system that embodies the newly understood tight interdependencies among a firm's functional units. ERPs have offered much promise for revitalizing IT infrastructures and enabling global business process integration (O'Leary, 2000; Sauer and Willcocks, 2001), but as the research in this book demonstrates, implementation of these highly touted systems has regularly proven to be very expensive, and the rewards for implementation often appear to be elusive.

A Mckinsey report looking at the 1995–2000 period concluded that spending on IT in the USA had doubled, as had the labour productivity annual growth rate over the same period. However, despite claims to the contrary, little of that productivity growth could be attributed to IT, and in fact the majority of the US economy saw no or little productivity growth while accounting for 62% of the IT spending in that period.

Introduction

Enterprise Resource Planning (ERP) systems purport to integrate the complete range of business processes and functions into a single information and technology infrastructure. By 2003 ERP software was highly configurable, and, while noting the qualifications made in earlier chapters, technically it went a long way towards accommodating the diverse needs of users across most sectors of the economy. This development has, in fact, happened over a long period. Though the breadth and tight integration of ERP has only become available in recent years, ERP have a pedigree in large, packaged application software that has been in widespread use since the 1970s (Klaus, Rosemann, and Gable, 2000).

The market for ERP products and services sustains a significant segment of the software and services industry, consisting of several of the largest software firms as well as the world's largest management consulting companies. Related annual revenues were expected to exceed $20 billion by 2002, approximately half vendor service revenue and half vendor license revenue (Gartner Group, 1999). Most multinational companies as well as many public sector organizations, have adopted ERP and increasingly small and medium-sized enterprises have been following suit. Indeed, by 2003 ERP formed the backbone of many application architectures and were being extended into developments such as e-business and e-government (see also Chapters 4 and 18).

Introduction

Problems associated with software implementations are not new, nor specific to enterprise resource planning (ERP) systems. Nevertheless, ERP systems have been blamed for the poor performance of several organizations (Osterland, 2000). A number of companies have reported negative impacts on earnings as the changeover to an ERP takes place, among them Hershey (Girard and Farmer, 1999), AeroGroup (Asbrand, 1999), and Snap-On (Hoffman, 1998). Hershey, for example, suffered a third quarter (1999) sales decrease of 12.4% and an earnings decrease of 18.6% compared with the preceding year (Osterland, 2000).

Implementing an ERP is a major undertaking and the Standish Group International estimates that 90% of SAP R/3 (the dominant ERP) projects run late (Williamson, 1997). The trade press is now also replete with articles on ERP failures, cancellations, and cost/time overruns. Dell canceled its R/3 system after two years when it determined that it could not support the required processes (King, 1997). AeroGroup (Asbrand, 1999), Unisource (IW, 1998), and a number of garbage disposal companies (Bailey, 1999) also abandoned their ERP projects. The most dramatic example of an organization claiming to be damaged by an ERP implementation is that of FoxMeyer Drug Corporation. FoxMeyer claims that its SAP R/3 system sent the company into bankruptcy (Bulkeley, 1996). Note, however, that, at the same time, many companies have experienced benefits that far exceeded their expectations (Davenport, 1998; Deloitte Consulting, 1998).

Introduction

Implementation of ERP systems is widely recognized (Parr, Shanks, and Darke, 1999; Ambrosio, 1997; Fine, 1995; Gartner Group, 1998; Horwitt, 1998; Stedman, 1998a, 1998b; Martin, 1998; Piszczalski, 1997; Tebbe, 1997) as both problematic, and likely to overrun time and budget allocations for the actual implementation project. Of course IT projects do have a history of having such problems but ERP implementations are more intractable than most. Martin (1998) reported that over 90% of ERP implementations were late and/or over budget; some have terminated in expensive legal actions (James, 1997; Shanks et al., 2000); and the popular literature (Infoweek, Computerworld, etc.) contains numerous stories which describe ERP implementations in colourful terms, such as ‘endurance’ tests, ‘fiascos’, ‘living to tell about it’, and ‘war stories’ (see, for example, Tebbe, 1997; Horwitt, 1998).

In response to these problems, the academic literature has taken a two-pronged approach. First, it has focused on CSFs (Parr, Shanks, and Darke, 1999; Holland, Light, and Gibson, 1999; Sumner, 2000; Shanks et al., 2000; Parr and Shanks, 2000b) for implementation of ERP systems. Secondly, it has developed process models of successful implementation (Bancroft, 1996; Bancroft, Seip, and Sprengel, 1998; Markus and Tanis, 2000; Holland, Light, and Gibson, 1999; Parr, Shanks, and Darke, 1999; Ross, 1998). Both approaches aim to achieve more successful implementation and deeper insight into the process.

Introduction

The enterprise resource planning (ERP) movement has been gathering momentum for the best part of ten years and has now reached a global dimension, with companies across the world and, more importantly, across very different industries jumping on the ERP band wagon. The pace of implementations has been such that SAP have posted on their web site the news that they have now implemented their software in 30,000 sites and have a user population of 10 million. This gives an idea of the scale and pace of the whole ERP movement.

ERP systems are integrated enterprise-wide software packages that use a modular structure to support a broad spectrum of key operational areas of the organization. They are widely acknowledged as having the potential to radically change existing businesses by bringing improvements in efficiency and the implementation of optimized business processes (Rowe, 1999). The key reasons why managers have sought to proceed with difficult ERP projects have been reported to be to end the fragmentation of current systems, to allow a process of standardization, to give more visibility on data across the entire corporation, and, in some cases, to obtain competitive advantage. Thus, ERP projects have been described as strategic projects whose success or failure will have a great impact on the organization (Rowe, 1999; Shakir, 2000; Wood and Caldas, 2000).

Introduction

One of the most enduring research topics in the field of information systems is that of systems success (Ballantine et al., 1996; DeLone and McLean, 1992; Lyytinen and Hirschheim, 1987). Prior research has addressed the measurement of success, the antecedents of success, and the explanations of success or failure. Yet, with each new type of information technology or application, the question of success comes up again. In the case of enterprise resource planning (ERP) systems, success takes on a special urgency since the costs and risks of these massive technology investments rival their potential payoffs. Failures of ERP system implementation projects have been known to lead to organizational bankruptcy (Bulkeley, 1996; Davenport, 1998; Markus and Tanis, 2000).

Briefly, ERP are commercial software packages that enable the integration of transactions-oriented data and business processes throughout an organization. From a base in manufacturing and financial systems, ERP systems may eventually allow for integration of inter-organizational supply chains (Davenport, 1998; Markus and Tanis, 2000). Because these systems touch so many aspects of a company's internal and external operations, their successful deployment and use are critical to organizational performance and survival.

Introduction

Companies continue to move away from developing IT systems in-house and purchase instead standard package software. PriceWaterhouse (1995) predicted that by the year 2000, two-thirds of all business software would be bought off the shelf. Following this Deloitte and Touche (1997) stated that enterprise resource planning (ERP) systems, standard package-based software that supports core business processes, were the preferred method by which businesses replaced ‘legacy systems’. In 2000 a Consultant's Advisory survey indicated that 61% of respondents planned to invest in existing or new ERP implementations. The use of standard systems has been the predominant strategy for several years in the area of desktop computing and it is clearly now being adopted elsewhere in organizations.

Even in the light of the phenomenal uptake of this form of strategy and subsequent potential for learning, implementation is still problematic for many organizations. There are mixed reports concerning the outcome of ERP projects. Successful ERP implementations are certainly publicised, such as Monsanto (Edmondson, Baker, and Cortese, 1997) and Guilbert-Niceday (Gibson, Holland, and Light, 1999), but less-successful attempts, such as FoxMeyer Drug (Bicknell, 1998; James, 1997), have received limited attention. It is estimated that at least 90% of ERP implementations end up late or over budget (Martin, 1998). However, this may be due to poor cost and time estimation rather than a failure in project management. Changes in the scope of ERP projects may also contribute to this figure (Holland and Light, 1999).

Introduction

Knowledge transfer is an important topic in management literature. Some studies have focused on the transfer of best business practices in multinational corporations as they transfer accumulated knowledge from their headquarters to new foreign affiliates (Ghoshal and Nohria, 1989; Gupta and Govindarajan, 1991; Kogut and Zander, 1993). Other studies have focused on the transfer of technology, business processes, and best practices inside a firm (for example, Argote, 1999). However, current studies have not focused on the unique knowledge transfer practices involved in an Enterprise Resource Planning (ERP) implementation.

ERP is an enterprise-wide application software package. In ERP, all necessary business functions, such as financial, manufacturing, human resources, distribution, and order management, are tightly integrated into a single system with a shared database. While customization is not impossible, the broad scope and close connectivity of all related functions make customization very costly (Davenport, 1998; Davis, 1998). The high cost and long implementation process of customization result in most organizations needing to align their business processes with the functionality provided by the ERP program rather than customizing the ERP package to match their current processes. According to Forrester Research, only 5% of organizations, among Fortune 1000 companies that had purchased an ERP application, customized it to match their business processes (Davis, 1998). Therefore, the implementation of ERP entails using the business models included in the package (Slater, 1998). In other words, business knowledge incorporated in the basic architecture of the software is transferred into the adopting organization.

Risk Factors in Enterprise-wide/ERP Projects

In the past several years many organizations have initiated enterprise-wide/ ERP projects, using such packages as SAP, PeopleSoft, and Oracle. These projects often represent the single largest investment in an information systems project in the history of these companies, and in many cases the largest single investment in any corporate-wide project.

These enterprise-wide/ERP projects bring about a host of new questions, because they represent a new type of management challenge. The management approaches for these projects may be altogether different from the managerial approaches for traditional MIS projects. Some of these questions and issues are:

• What are the major risk factors associated with implementing traditional MIS projects?

• What are the major risk factors associated with enterprise-wide information management projects?

What are the differences? What new risk factors need to be addressed in ERP projects? What are some of the risks in ERP projects that are not factors in non-ERP projects?

Most organizations have extensive experience managing traditional MIS projects, but these new ERP projects may represent new challenges and present new risk factors that must be handled differently. This paper will provide case studies of seven organizations implementing enterprise-wide/ERP projects and will provide insight into each of these questions based upon their experiences.

Introduction

As we look back over the contents of this book, one thing becomes very clear. The Introduction's claim that one of the most significant IT developments in recent years has been the widespread implementation of ERP packages by large and medium companies has been substantially justified. Adopting packaged software allowed many companies to replace their ageing legacy systems in time to avoid Y2K problems and brought a variety of other benefits, including strategic business advantages, improved system architectures, and outsourced software maintenance. With so much going for them, ERP packages seem likely to remain popular for some time to come. In this context, it makes sense to build on the picture painted in the Introduction, and anticipate how ERP packages will evolve in the future.

One view of the future can be constructed from an assessment of the issues faced by today's ERP package adopters and the responses under development by ERP package vendors. In this view, the functionality of ERP packages will expand, and the architecture of ERP packages will evolve, in ways that address many of the 2002–2007 business opportunities and technical challenges. This view of the future assumes a high degree of continuity with today's in-house IT management regime; in particular, it involves a continuation of the current division of labour between ERP package adopting organizations and ERP package vendors (Brehm and Markus, 2000). A clear articulation of this continuity view of the future is offered by Davenport (Davenport, 2000a).

Introduction

ERPs – software packages that manage and integrate business processes across organizational functions and locations – cost millions to buy, several times as much to implement, and necessitate disruptive organizational change. While some companies have enjoyed significant gains, others have had to scale back their projects and accept minimal benefits, or even abandon implementation (Marcus and Tanis, 2000).

Historically, a common problem when adopting package software has been the issue of ‘misfits’, that is, the gaps between the functionality offered by the package and that required by the adopting organization (Davis, 1988; Lucas, Walton, and Ginzberg, 1988). As a result, organizations have had to choose among adapting to the new functionality, living with the shortfall, instituting workarounds, or customizing the package. ERPs, as a class of package software, also present this problematic choice to organizations.

Even though there are many built-in switches that one can manipulate to customize the software, smooth alignment of the software functionality to business requirements is still far from ideal as many of these issues have to be dealt with at the application architecture level. AeroGroup, for example, has dropped the Apparel Footwear Solution of SAP/R3 because of its inability to model the uniqueness and complexities of the footwear business. Similarly, the failure of Fox Meyer could also be partially attributed to the adoption of an ERP system that was designed more for manufacturers rather than wholesale distributors.

Introduction

By 1999 the ERP ‘revolution’ was generating over $US20 billion revenues annually for suppliers and a further $20 billion to consulting firms. If 2000 and 2001 saw a fall-off in business globally, then some believed that once the distractions of Y2K deadlines were over, new ERP business plus the need to support and capitalise on the sunk investment in IT infrastructure ERP already represented would guarantee further take-off. Once again, at first without many noticing, information technology (IT) not only raised itself high above the cost parapet, but also set off traditional alarm bells about questionable business value. For many firms ERP represents the return of the old IT catch 22 with a vengeance – competitively and technically it is a ‘must-do’, but economically there is conflicting evidence, suggesting it is difficult to cost justify, and difficult to derive benefits from.

The problem has been further complicated by the coming of web-based technologies. Thus Sauer and Willcocks (2001) found many organizations struggling to integrate their legacy and relatively new ERP systems with e-business initiatives and technologies. The goal of a relatively seamless e-business infrastructure seemed particularly difficult to achieve in ‘bricks and mortar’ companies trying to move to the web (Sauer and Willcocks, 2000), a finding explained by Kanter (2001) and Willcocks and Sauer (2000) in terms of cultural, organizational, and political issues, together with less than good organization and project management for e-business.