MTU Library Catalogue

Syndetics cover image
Image from Syndetics

Mission critical : realizing the promise of enterprise systems / Thomas H. Davenport.

By: Davenport, Thomas H, 1954-.
Material type: materialTypeLabelBookPublisher: Boston, MA : Harvard Business School Press, 2000Description: x, 335 p. : ill. ; 25 cm.ISBN: 0875849067.Subject(s): Management information systemsDDC classification: 658.4038
Contents:
What are enterprise systems and why do they matter? -- The promise and perils of enterprise systems -- Should my company implement an enterprise system? -- Linking enterprise systems to strategy and organization -- Linking enterprise systems to business processes and information -- Achieving value during enterprise system implementation -- Transforming the practice of management with enterprise systems -- Using enterprise systems to manage the supply chain -- The future of ES-enabled organizations.
Holdings
Item type Current library Call number Copy number Status Barcode
General lending MTU Bishopstown Library Lending 658.4038 (Browse shelf(Opens below)) 1 Available 00160634
Total holds: 0

Enhanced descriptions from Syndetics:

This is a no-nonsense guide to the benefits and pitfalls of enterprise-wide information systems. How many organizations would doubt the promise of an integrated enterprise system (ES)? Not many, judging by a $15 billion industry. The combination of an ES as a platform for organizational information and Internet technology for gaining access to it adds up to the ideal solution for company-wide data sharing in real time. Not surprisingly, small and large companies worldwide are either considering an ES, in the process of implementing one, or living with the results. Yet, says Tom Davenport, unless managers view ES adoption and implementation as a business decision rather than a technology decision, they may be risking disappointment Mission Critical presents an authoritative and no-nonsense view of the ES opportunities and challenges. Suggesting ESs are not the right choice for every company, the author provides a set of guidelines to help managers evaluate the benefits and risks for their organizations. To be successful, argues Davenport, an organization must make simultaneous changes in its information systems, its business processes, and its business strategy. Such changes are described in detail with extensive examples from real organizations. Bolstering his contention that ESs should be viewed as business vs. technology projects, Davenport spells out the specific business change objectives that should be formulated in advance of ES adoption and monitored throughout its implementation. The first strategic guide to the ES decision, Mission Critical will be indispensable to general managers and information technology specialists at all stages of the implementation process.

Includes bibliographical references (pages 313-320) and index.

What are enterprise systems and why do they matter? -- The promise and perils of enterprise systems -- Should my company implement an enterprise system? -- Linking enterprise systems to strategy and organization -- Linking enterprise systems to business processes and information -- Achieving value during enterprise system implementation -- Transforming the practice of management with enterprise systems -- Using enterprise systems to manage the supply chain -- The future of ES-enabled organizations.

Table of contents provided by Syndetics

  • Preface (p. vii)
  • Acknowledgments (p. ix)
  • 1 What Are Enterprise Systems and Why Do They Matter? (p. 1)
  • 2 The Promise and Perils of Enterprise Systems (p. 29)
  • 3 Should My Company Implement an Enterprise System? (p. 55)
  • 4 Linking Enterprise Systems to Strategy and Organization (p. 105)
  • 5 Linking Enterprise Systems to Business Processes and Information (p. 135)
  • 6 Achieving Value during Enterprise System Implementation (p. 169)
  • 7 Transforming the Practice of Management with Enterprise Systems (p. 203)
  • 8 Using Enterprise Systems to Manage the Supply Chain (p. 237)
  • 9 The Future of ES-Enabled Organizations (p. 265)
  • Appendix A Technical Overview of Enterprise Systems (p. 299)
  • Notes (p. 313)
  • Index (p. 321)
  • About the Author (p. 335)

Excerpt provided by Syndetics

Chapter One WHAT ARE ENTERPRISE SYSTEMS AND WHY DO THEY MATTER? AROUND THE GLOBE COMPANIES ARE QUIETLY AND STEADILY becoming more connected--one business function with another, one business unit with another, one company to another. They are putting information systems in place that will yield more and better information more quickly than they have ever known. For the first time since large businesses were created, managers will be able to monitor the doings of the company in near real time, without having to wait for monthly reports that must be cross-referenced with other monthly reports, all of which may be out of date--or just plain wrong--by the time managers receive them. Managers will also be able to sit in front of their workstations and know what is happening at every point around the globe.     For the first time ever, information will flow seamlessly across diverse business functions, business units, and geographic boundaries. What the Internet is doing for communications between organizations, these systems are doing within companies. For better or worse, no business transaction--no customer purchase, no supplier invoice, no product produced--will go unnoticed by these systems. Ultimately, every bit of computer-based information used for running a company's operations can be supplied by these systems. This situation sounds utopian, but it's actually available today if companies can master a relatively new type of information system.     Let's call such information systems enterprise systems (ESs). Also known as enterprise resource planning (ERP) systems, these are packages of computer applications that support many, even most, aspects of a company's (or a nonprofit organization's, university's, or government agency's) information needs. The ERP name reflects the manufacturing roots of these systems--it's a modification of "MRP" (manufacturing resource planning)--but in my view these systems have so transcended their origins that the somewhat clumsy ERP name is no longer appropriate.     From accounting to manufacturing, from sales to service, ES modules support thousands of business activities. Aside from personal productivity applications such as spreadsheets and word processors on personal computers, highly specialized production systems such as process control, and Internet-based systems for information and knowledge access, an ES may be the only business information system an organization requires. This breadth is one of the key factors that distinguishes enterprise systems from earlier systems.     Enterprise applications started out as "back-office" systems, automating the workmanlike business transactions that customers never saw or cared about. Sure, ledgers needed to be updated, suppliers paid, and employee vacation balances debited, but accuracy and speed in this type of transaction rarely led to any competitive advantage or increased customer satisfaction. Although back-office systems may not offer competitive advantage, they do have important implications. Poorly functioning back-office systems can lead to dissatisfied customers, suppliers, auditors, or regulators. If a company can't generate an accurate invoice, meet a promised delivery date, find a missing shipment, or properly account for costs and revenues, it can get in big trouble quickly. Well-implemented enterprise systems can make well-executed back-office transactions a reality.     More recently, however, ESs have moved into the front office, supporting supply chain optimization, sales force automation, and customer service. These new functions have been achieved either by installing more comprehensive packages from ES vendors or by installing complementary software applications--sometimes called bolt-on systems --from third-party software companies. Because the goal is to have added capabilities connect smoothly with the core ES system, I'll refer to the entire entity as an ES, even though it may consist of several different components.     Even more recently, a new technology has largely erased the distinction between front and back offices. The Internet, and associated internal networks called intranets , is the ideal tool for distributing and providing access to information. With just a browser, employees, suppliers, and customers can all access the organization's information. But where will that information come from? Internet technology itself is not suited for processing business transactions or for storing key data--it's an information access technology. Enterprise systems are, of course, perfectly suited for information transactions; they're the underlying information factory producing the information for internal and external Internet consumption. Using the Internet to give employees and customers access to poor-quality, unintegrated information is like opening more bank branches when the currency is worthless. You have to work on both access and high-quality information simultaneously. The combination of enterprise systems as the primary platform for organizational information and of Internet technology for providing access to it will be the hallmark of leading organizations in the new century.     Whether front office or back office, by themselves or in combination with other technologies, ESs are distinguished by their information commonality and integration. It's great to have the entire business supported by a single type of information system, but what if the information differs from one part of the organization to another? For the most part, this is prevented in ESs through the use of a common database for the entire organization. Not only can one track customers through marketing, sales, and service activities, but the customer's identification number and address are constant across those different applications and business functions. The Babel-like information environments of most large organizations, in which the same term might mean different things in different parts of the company, can be avoided altogether through the use of an ES.     In short, these systems offer just about everything businesses want from a computer. They serve up information in a format that anyone--not just technologists--can understand. They employ client/server technology--the state of the commercial art in information systems. They even work well with the Internet. If these systems are so good, why wouldn't every organization want one?     In fact, they do. With a few exceptions I'll discuss later, large, medium-sized, and increasingly even small organizations are installing enterprise systems: from Iowa Spring, which has about $11 million in revenues, to its customer General Motors, which is more than ten thousand times larger. Public-sector organizations ranging from the "city" of Round Rock, Texas, to the Victoria Department of Education in Melbourne, Australia, have them. In some industries, such as petrochemicals, every company has an ES. In others, such as the electrical utilities business, ESs are being adopted at a rapid rate. Even in financial services, one of the industries in which these systems have been less popular, hundreds of organizations from Bank One to Dai-Ichi Life Insurance have them in place.     The software and hardware spending alone for ESs is well over $15 billion per year worldwide, and professional services fees add another $10 billion. Revenues for ES vendors have grown between 50 and 100 percent a year. Very large companies, such as Hewlett-Packard, Procter & Gamble, and Intel, speculate that their ES expenditures will easily top $1 billion before they are finished. (As I describe later, they will never be finished, so their total costs will ultimately be even higher!) In such firms the cost of implementing an ES is orders of magnitude higher than, for example, the money spent on the Internet, Web sites, and electronic commerce.     Of course, with every benefit there is risk. A few companies have failed at implementing ESs; many more have spent more than they intended or encountered resistance from managers and workers who were unprepared for the changes ESs bring. Companies are willing and correct to take these kinds of risks because of the impact the systems can have on quality, cost reduction, and customer satisfaction and loyalty.     Enterprise systems offer the first great opportunity to achieve true connectivity, a state in which everyone knows what everyone else is doing in the business all over the world at the same time. And because they represent the first great opportunity for connectivity, they pose one of the greatest threats to the status quo that companies have ever faced. Because companies are made up mostly of people, ESs mean you will have to change people and the way they do things at the same time that you change all the computers and the software. That is why ESs may be more rewarding--and more challenging--than any computer system a company has ever tried to install. Big Systems, Big Change Being successful with enterprise systems is not simply a matter of writing big checks. What's really important--and difficult--about these systems is the dramatic change they bring to a business. I'll argue throughout this book that an enterprise project is as much about changing the way a business operates as it is about technology. Successful implementation of an ES does involve probably the greatest technological change most organizations have ever undergone, not to mention the largest employing client/server technologies. Even more difficult and important, however, are the major changes in business that come with an ES project.     Business processes, the way work gets done in an organization, change dramatically. Organizational structure and culture, the behaviors of workers throughout the company, and even business strategy all have to be restructured. The reengineering movement of the early 1990s, with all its radical approaches to reorganizing companies, turned out to be a mere preamble to the ES era, which has brought even more ambitious (and complex) changes. In fact, the business process reengineering movement has largely been replaced by ES initiatives. Given their breadth and technical complexity, ES projects are even more difficult and consuming of time and resources than the largest reengineering projects. The most ambitious ES projects can take a decade or more of a company's time.     Implementing new mission-critical systems, then, is hardly just a matter of installing an ES. Business processes and information must be made common around the world within the implementing organization. Idiosyncratic ways of doing business must be abandoned. Informational linkages between business functions and units must be tightened. Employees must be educated about the broad implications of simple actions like pressing a key within an ES. Perhaps most difficult of all, senior managers must be persuaded of the wisdom of changing virtually everything in a company at once. In short, organizational change represents a huge part of a successful ES project. Managers at Steelcase, for example, estimate that up to half of the company's project resources went for organizational change issues. A Monsanto manager felt that change management activities constituted 75 percent of the total project effort there.     Despite these difficulties, ESs are the answer to the Information Age's wildest dreams. The concept of an integrated set of information technology (IT) applications that could meet all of an organization's information needs has been with us since the beginning of information systems in business, but has been unrealizable before the modern ES. We have gotten what we wished for; now we only need to make the business and organizational changes necessary to take advantage of our fulfilled dreams. Business Benefits of Enterprise Systems I've already pointed out that ESs are difficult to put in place from both a technical and business change standpoint. Why go to all the trouble and expense of implementing an ES? In an ideal world, ES-enabled organizations would be seamlessly interconnected both internally and externally. Excess inventory and waste would be nonexistent. Demand and supply would be perfectly coordinated. It would be just as easy to transact business with suppliers and customers as with another department of your own company. Customers would have perfect information about not only the products and services they've ordered from you, but also about how every aspect of your business affects theirs. Managers could understand any aspect of a company's operations and performance with a few clicks of the mouse.     These benefits aren't purely hypothetical. Many companies have already realized substantial business benefits from their ES projects, even if they are not completely finished installing them. Several examples of these benefits are described in the following paragraphs. 1. Cycle time reduction. Autodesk, a leading manufacturer of computer-aided design software, has achieved substantial benefits in terms of cost and time reductions in key business processes. Whereas the company used to require two weeks on average to ship to customers, 98 percent of products are now shipped within twenty-four hours. Financial closing times were cut in half, from twelve days to six. Autodesk calculates that it has saved more on reduced inventory alone than its SAP system cost to install. 2. Faster information transactions. IBM's System Storage (disk drive) division achieved a reduction in the time to enter pricing information from five days minimum to five minutes, replacement part shipping went from twenty-two days to three, and credit checks that previously took twenty minutes are now accomplished in three seconds. Crediting a customer for a returned disk drive used to take three weeks; it now happens immediately. The division once spent thousands of hours reconciling management reporting data; this now happens automatically. IBM in general has twenty-one SAP projects underway, covering 80 percent of its core business; eight projects are up and running. 3. Better financial management. Microsoft is installing an ES to bring about common financial and procurement systems worldwide. The fast-growing software company has already saved $2 million in equipment depreciation (it previously took three months to start the depreciation schedule for a new asset; now it can begin immediately). The company's ES has allowed it to receive $14 million per year in early-payment discounts from vendors. Microsoft's managers also report substantial benefits in improved management and reporting systems, and the financial closing cycle has been reduced from twelve days to four. 4. Laying the groundwork for electronic commerce. Cisco Systems put in an ES to structure and rationalize its back-office business transactions systems, which were previously unable to support the company's rapid growth. Without the system, Cisco also wouldn't have been able to offer customers Web-based access to product ordering, tracking, and delivery processes. Cisco's system cost it over $15 million, and the company spent another $100 million connecting it to the Internet. Today, however, Cisco believes that the combination of its ES and its Internet applications yields more than $500 million in annual operating cost savings. 5. Making tacit process knowledge explicit. Monsanto was concerned that decades of knowledge about plant operations existed only in the minds of an aging workforce. After successfully implementing its ES, company managers now feel that key processes, decision rules, and information structures are well understood and documented in its system. Furthermore, the knowledge is now more common to the industry, so that new employees are more likely to understand the work process. Support of the process and the system can also be outsourced to external suppliers.     The primary lesson from these examples is that key business processes can be improved dramatically through the implementation of an ES. Whether the process is financial, managerial, or operational; whether it involves internal activities or customers and suppliers; whether the process runs faster or leaner--enterprise systems are the primary vehicles for making business processes better. It's virtually inconceivable to try to reengineer today without them. Business Life before Enterprise Systems In order to better understand the value of ESs, it's useful to contrast them with the way that organizations previously met their information needs. In 1954, when the first business application of computers was developed (by what is now Andersen Consulting for General Electric), and for most of the following forty years, when a business function needed computerized information it used a stand-alone application. The first application created was for payroll processing; later ones would be created for general ledger, accounts payable, inventory management, or customer billing. Each system had its own application logic, its own information, and its own user interface. An individual company might have hundreds of individual applications. Even when software vendors began to sell application packages of broader functionality in the 1980s, they were almost always within individual business functions, for example, finance and accounting. We haven't totally left this approach behind today. Companies that don't have an ES still have a variety of standalone systems.     Chopping up information systems this way makes it impossible to coordinate planning across different business functions. Say, for example, that a company wants to compare information from its manufacturing and sales functions so that it doesn't produce more inventory than it can sell. The information is there within the company, but it's not accessible or comparable--making things frustrating for those who need it. In most cases in the past, this coordination simply couldn't be done in any automated fashion because companies' manufacturing systems were separate from their sales systems. The sales force didn't know what manufacturing had produced recently, and manufacturing didn't find out until later what the sales force had sold. The idea of "available to promise" inventory (i.e., stuff a company has made or could make that isn't promised to anyone else, so is available to be sold) just didn't exist. Connections between functions--and often between different geographical areas--were loose and slow. When connections did take place they involved many middle managers, whose jobs entailed collating and passing this information around the organization.     The most insidious aspect of this problem involved different interpretations of the same information entities. The term customer , for example, might appear in many of these disparate systems around a firm. In one system it might include distributors, in another only end customers. In one version it would incorporate prospects, in another only existing customers. If the CEO asked for a list of the top 100 customers, it could take weeks to come to consensus on the list (as happened at one computer company with diverse systems around the organization). Having diverse forms of information has some positive attributes as well (e.g., every part of the organization gets the interpretation of customer that best suits its needs), but there is no doubt that the proliferation of information meanings can cause considerable confusion. One department's interpretation of how much has been sold through what channels may differ from another's. Every department could have its own interpretation of how much money was passing through based on what information they each believed described a customer. Obviously, this disparity made it difficult to make good decisions about which customers to serve, which selling approaches were most effective, and whether to build the business or hold back.     Having multiple systems that could not talk directly to each other was--and still is for those organizations who haven't moved fully to ESs--a maintenance nightmare. Managing hundreds of different systems means managing tens of computer languages, hundreds of different maintenance and update schedules, and thousands of pieces of documentation. In most organizations there is literally no one who understands how all of the pieces fit together. When one system needs to be connected to another, ad hoc connections must be made, the maintenance of which becomes another problem. Because of these difficulties, many companies spend more than half of their information systems budgets on maintenance. Managing communications between computer systems that were never designed to talk to one another requires enormous, continual work on interfaces. When any of the systems is changed, all of the interfaces have to be changed as well.     If a major problem occurs in these multiple standalone systems regarding system design or functionality (say, purely hypothetically, of course, that the date field in many systems held only two digits as the year 2000 was approaching), finding and correcting the problem across so many systems is extremely difficult. In fact, the Year 2000 (Y2K) problem has been a major driver for many companies to install an ES. Companies adopted a "kill two birds with one stone" approach, solving their Y2K problems while installing a more functional and integrated system. Current versions of ESs from major vendors can all deal with 2000 and subsequent years (at least until the year 10,000!). Another example of this "simplification through integration" approach is conversion to the new common European currency, the euro. Those organizations with multiple nonintegrated systems need to incorporate the new currency across each system; companies with an ES need only install one system (or a new release of their existing ES) that can handle the euro.     It is possible, of course, for companies to build proprietary ESs just as they have written proprietary standalone software programs. A few companies have succeeded in this regard. VeriFone, for example, which provides banks and retailers with transaction automation via its card-swipe devices, has long had its own fully integrated system for internal business transactions using a common database. VeriFone's managers and programmers designed and built the system, and it works well. However, the success rate for companies trying to build their own systems is quite low, and the price quite high. Many large banks, for example, have attempted to build integrated systems only to fail. Even at VeriFone, the company's system runs on an obsolete computing platform and needs to be rewritten. Further, the company has been acquired by Hewlett-Packard, which is implementing an ES from a large vendor. Hence the future of VeriFone's home-grown ES is in doubt. Most organizations should not even consider developing their own system. Few companies do business in a unique enough fashion to benefit from such a move. Business Life Today with Enterprise Systems With ESs, you don't have to build your own integrated system. You can buy it from one of several vendors. Each vendor offers more or less the same overall product: a set of application modules that all fit together. Each module includes a variety of functions; for example, the accounting module from almost every vendor includes general ledger, accounts receivable and payable, funds management, financial consolidation and reporting, foreign exchange, and cash management. Most of the information produced by these modules is already Internet or intranet accessible; all of it will be before long.     A company need not install all possible modules. Some modules (e.g., those for finance and accounting) are implemented by almost all firms; others (e.g., human resource management) may not be included in a particular ES project. The company may already have serviceable systems in that part of the business, or it may choose to use standalone "best of breed" functionality rather than a somewhat less functional ES module that is fully integrated with other systems and business processes. The greater the number of modules selected, the greater the integration benefits, the more need for business change, and the higher the cost and the risk of the implementation project.     Companies assemble their choices of modules and install them as a complete system, perhaps adding one or more additional applications from third-party vendors. All of the applications work with the same data, defined in the same way and stored in a common database. A business transaction recorded in one application ripples through the entire system, and all relevant data values are updated.     For example, let's assume that a South Africa-based sales-person for a U.S. multinational computer firm prepares a quote for a customer using an ES. The quote specifies a legally binding product configuration, price, delivery date, shipping method, and so forth, all determined in real time from the system. When the customer accepts the quote (via the Internet, let's say), a sales order is recorded. The system schedules the shipping (including shipping points and truck routes), then works backward from the shipping date to reserve material availability, order needed parts from suppliers, and schedule the computer assembly in manufacturing. The customer's credit limit is checked. The sales and production forecasts are updated. MRP and bill-of-materials lists are created. The salesperson's payroll account is credited with the correct commission, and his or her travel account credited with the expenses of the sales call. Actual product cost and profitability are calculated. The division's and firm's balance sheets, accounts payable and receivable ledgers, cost center accounts, cash levels, and any other relevant financials are automatically recalculated in an instant. Virtually every information transaction resulting from the sale of the computer is taken care of except for the impact on the company's stock price (alas, ESs cannot yet calculate investor psychology). A Key Choice in Implementing Enterprise Systems All ESs work in roughly the manner just described, but companies can implement them in several different ways. The two key dimensions that differentiate approaches to ES implementation are the time it takes to implement, and the amount of business change and value to which a company aspires. These dimensions, when combined, form the matrix of approaches in figure 1-1.     Enterprise systems can be implemented quickly or slowly, depending on how ambitious the company's goals are, how pressing any deadlines are, and how well implementation proceeds. A fast implementation might take as few as six months; a slow one can take up to five years or more. Enterprise systems can be installed for technical reasons or to enhance strategy and competitiveness. A technically focused implementation is intended only to provide core information systems functionality to an organization, with as little business change as possible. A strategic implementation attempts to maximize positive business change and business value.     The only combination of these dimensions that it makes no sense to strive for is the slow, technical option. Since the technical focus provides little business value, it makes sense to complete it as quickly as possible. If you find yourself in this category, it probably means you've just experienced a poor implementation.     A fast, technical implementation can mean quick relief from pressing technical problems or inadequate legacy systems. This approach is probably the least expensive implementation option. It provides no direct business value other than removing any technical barriers to business effectiveness. Some companies say that they are starting with a fast, technical approach but are planning to later achieve significant business value from their systems. ("We'll put the system in quickly, and then later on optimize it to fit our business.") I wonder just how many will ever get around to the optimization phase. It's too early to say whether this is a valid approach, but I don't really advocate it unless the organization is truly threatened by its technical problems.     Instead, I advocate that companies strive for business value in their ES implementations. It's only logical that when a company spends the millions it takes to put in an ES it should try to achieve as much business value as possible from the project. And evidence from an Andersen Consulting survey of 200 CEOs of companies that have implemented ESs suggests that projects viewed primarily as business initiatives are more likely to be perceived as leading to satisfactory business results. Fortunately, 85 percent of the CEOs in this survey, at least, did view their ES projects primarily as business initiatives.     In an ideal world, companies could transform their strategies and enhance overall competitiveness while completing ES implementations within a few months. However, the comprehensively strategic and fast project probably doesn't exist. It is possible, however, to adopt a quick approach and achieve some competitive differentiation. This simply means that the implementation must be heavily focused on a specific business process or means of competitive advantage--for example, the supply chain or customer service. Most of the remaining business processes would have to be viewed in the same terms as when considering a fast, technical implementation.     Companies don't like to admit that their implementations are slow, but it may be worth a drawn-out implementation if it leads to considerable business benefit or competitive differentiation. In fact, if you're looking to change a broad range of processes, improve the way you relate to customers and suppliers, create a new organizational culture, and modify the behaviors of many workers, it's going to take a long time. Don't worry about it. Criticisms of Enterprise Systems From both the technical and business perspectives, ESs have their faults. If you are trying to decide whether to implement one, you should be aware of both the opportunities they provide and the problems they portend. Even if you've already committed to an ES, you should read this section because you may discover difficulties that lie down the road. On balance, however, I believe the criticisms can be overcome and that some ES is a good fit for most large organizations. Inflexibility. Some would argue that today's enterprise software offerings are too inflexible. It is just too difficult, this viewpoint would have it, to fit an ES to a business--both for the first time and for subsequent changes. Further, many times companies end up doing business in a way that they don't really want just because the system requires it. Another aspect of this supposed inflexibility is that once an ES is installed in your organization, it's too difficult to change how you work and are organized. ESs are like cement, the critics say--highly flexible in the beginning, but rigid later.     There is some truth to these criticisms. One of the greatest difficulties in any ES project is to match the system to one's preferred ways of accomplishing a business process or activity. And most companies have only experienced this problem once; future updates of the system to meet changes in the business may be as difficult as the first go-round. Intel, for example, has already found that it needed twelve additional people to adapt its SAP system to day-to-day changes in its organizational and financial structure.     In response to this inflexibility charge, I ask companies "Compared with what?" A hypothetical object-oriented, highly modular system may someday provide greater flexibility than today's ESs, but no such system is available today, or even on the immediate horizon. Another answer to the flexibility issue is that some ESs are more flexible and easily modified than others. In general, there is a trade-off between the comprehensiveness and breadth of an ES package and the ease with which it can be configured and modified. Thus, organizations do have a choice, obviating some of the criticism. And it is certainly true that today's ESs are more easily configured than those of the past: ES vendors and third-party firms have both developed tools to help companied tailor a system to their businesses. Finally, there are a number of companies who have argued that putting in an ES actually made them more flexible in response to a changing business environment simply because they have only one system to change, not many. (Continues...) Copyright (c) 2000 President and Fellows of Harvard College. All rights reserved.

Author notes provided by Syndetics

Thomas H. Davenport is the Director of the Andersen Consulting Institute for Strategic Change and a Professor of Information Management at Boston University.