Bob Benson, CFO of telecommunications company Aliant Inc., crows that his company's three- year enterprise resource planning (ERP) project has come in on time and on budget. In 1997, four loosely allied telecom Þrms in Eastern Canada--Bruncor, Island Telecom, Maritime Telegraph & Telephone, and NewTel Enterprises--decided to form an information systems "co-op," uniting 36 different financial, project management, and human resources systems under a single PeopleSoft umbrella. The ERP project, now in its final phase of implementation, has "surpassed our business-case expectations," says Benson.
Indeed, cooperative projects like this one succeeded so well that, in June, the four companies decided to merge, becoming Aliant, a $2 billion company with 9,000 employees. Cost to implement the ERP system, to date: $20.2 million. Projected annual savings, in operating costs, from the system: $8 million. Anticipated return on investment (ROI): 33 percent.
Aliant determined its ROI by "conducting a detailed review of all our associated costs-- both current and proposed under the new system- -and, from that, calculating our savings," says Benson. "To arrive at tangible metrics, we involved users directly to discover what it costs to do the work they do now, and what it would cost once we got PeopleSoft implemented."
The on-time, on-budget arrival of Aliant's system is a rare occurrence in the world of ERP. In fact, it almost never happens, maintains Jim Johnson, chairman of The Standish Group, a research advisory firm in Dennis, Massachusetts. According to Standish Group research, 90 percent of ERP implementations end up late or over budget, and Johnson calls that estimate "conservative." No wonder that stories like Aliant's have been overshadowed by tales of ERP train-wrecks. System derailments at such companies as Kellogg's, Dell Computer, Dow Chemical, and Fox-Meyer Drug have made headlines over the past few years. "In the ERP world, there's lucky and there's lousy," says Johnson. "Aliant is one of the lucky ones."
If its ROI materializes as expected (and, in fact, the benefits are starting to accrue), Aliant will be doubly blessed. According to a 1999 survey of large multinational companies conducted by IT consultancy Meta Group, in Stamford, Connecticut, it takes an average of 23 months to implement an ERP system and another 31 months--four and a half years total- -to see demonstrable benefits from the effort. Over a five-to-six-year period, the average company incurred a negative net present value of $1.5 million, reports Meta Group. (Most organizations eventually realized positive cash flows associated with the ERP initiative, says Meta Group.)
What's more, the survey revealed that the total cost of ownership for a single ERP system--including hardware, software, training, support, and maintenance--runs at a stunning annual average of $15 million, or 2.4 percent of corporate revenues. (The smaller the company, the larger the percentage, the study found.)
To be sure, multinational, multisite ERP implementations are an order of magnitude more difficult than plunking down a single-site system in a small manufacturing company. Still, given the size of ERP projects, survey results like Meta's, and the horror stories, it's puzzling that more companies don't demand or can't produce a hard ROI from these huge investments. And though one can argue that it isn't easy to measure soft benefits with financial yardsticks, CFOs are finding that they'd better try anyway.
"Management-accounting principles for information-technology projects are well overdue," says Michael Pedersen, senior vice president and managing director at Meta Group. "But information technology is an asset like any other asset. CFOs need to be able to measure what they're getting from it."
When ERP is Self-Evident
Clearly, there are compelling reasons for ERP that have little to do with purely financial rationales. For instance, a company's legacy systems may be antiquated or non-Y2K- compliant. Growth and acquisitions may have created a poorly integrated hodgepodge of systems and databases. Or, the company's business model may have changed fundamentally. In such cases, insisting on a strict ROI may seem somewhat beside the point.
When Hollywood, Florida-based Memorial Healthcare System implemented its Lawson Software system two years ago, "We did not go out and do an ROI," recalls CIO Dennis Miller. Why? "We were sitting with a 20-year-old mainframe system designed for a hospital, and we had become an integrated [health-care] delivery system with multiple entities." And, like many mainframe systems, Memorial Healthcare's legacy programs also had the year 2000 problem. "We spent our time on evaluating the right system, not whether to buy one," says Miller.
The need for ERP was also self-evident at Millipore Corp., a $700 million Bedford, Massachusetts-based company that makes purification filters for the pharmaceuticals and microelectronics industries. Uniting Millipore's 31 offices around the world under a single Oracle-based system "was a necessity, not an option," says vice president and CFO Fran Lunger. "We could not serve our global customers if we had research information in one database, our general ledger in another, and manufacturing information in another."
Although Lunger won't say how much the company has spent on ERP as a cost of doing business, he is absolutely certain that the effort has produced clear benefits in addition to uniting the company under a single data center. Yes, there are immediate benefits: Millipore can close its books in four days. But more important are the less definable advantages: "At the end of the day," Lunger says, "we're servicing our customers in the way they want to be serviced. How do you measure that?"
Tangible, Intangible, and Strategic
Admittedly, even the term "return on investment" can be slippery. "Most people use the term 'ROI' to stand as a calculation to show the value of something, but it's also a code word for 'business justification,'" notes Jack Keen, president and founder of The Deciding Factor, a Basking Ridge, New Jersey based software development and consulting firm that builds ROI models. "Most people base their justifications on 'tangible' benefits alone. That's a mistake."
Generating a positive ROI is a multipronged task, say experts. It involves analysis to help manage companies' expectations, by revealing where the hidden costs and returns on ERP really lie. It requires carefully sticking to project parameters, and it means not being afraid to put price tags on "intangible" things.
Eran Grumberg, director of Meta Group's enterprise package services, says ERP systems can yield multiple benefits, which he classifies as "strategic," "tangible," and "intangible." Traditionally, he says, CFOs have used only one of these three yardsticks-- the "tangible" ones, such as lowering head count or shrinking inventory--and have thus missed out on possible returns.
To be sure, a "strategic" benefit--something that's of long-term value to the company, such as growth in the company's market share (part of which might be attributable to the company's ERP system), may not be measurable, as Millipore's Lunger suggests. And certainly, measuring an intangible benefit like "customer satisfaction" isn't easy. But that doesn't mean it's impossible, or that the effort shouldn't be made.
Indeed, many if not most "soft" benefits can correlate to hard numbers, provided project analysts are willing to do some imaginative number-crunching. Faster and more accurate product-delivery times are key to customer happiness. "Cleaner" information can result in fewer costly errors. In some cases, the ability to close books faster may be a tangible benefit disguised as an intangible one. Discovering the tangible benefit requires an exercise in analysis--breaking business processes and jobs into fine-grained descriptions.
Jack Keen recalls one instance in which the CFO of a multibillion-dollar company asked him to help cost-justify his company's ERP investment. "I asked him, 'How much money could you save by closing the books faster every month?'" Keen recalls. "To him, 'closing the books faster' felt like an intangible. But we asked field sales managers what they might save on a per-day cost basis if they didn't have to hire extra personnel due to out-of- date sales forecast data. It came to a million- dollar benefit over five years. If we hadn't brainstormed on that, we would never have found that million dollars."
To help his clients convert intangibles to tangibles, Keen licenses them tools he calls "flashcards." Each of Keen's 1,000 flashcards deconstructs an intangible ERP benefit through a formula for calculating current costs and estimated savings costs over time, beginning on the first day of the fiscal year.
Of course, it takes more than a set of flashcards to develop a complete ROI calculation for an enterprise system. Meta Group's Pedersen says CFOs should treat ERP as a line of business unto itself, with its own detailed business plan and a separate set of books. "CFOs need to be rigorous about treating this in the same way they would any other capital-allocation function," he insists.
That means building a careful plan with detailed definitions, then managing the project to the plan, adds Debra Hofman, a research director with Benchmarking Partners, Inc., a research and consulting firm. "If the goal is a 25 percent reduction in inventory, then you need to build a plan that spells out exactly what it takes to do that, and drive to that plan," says Hofman.
ROI is in the Details
Aliant's Benson agrees. He credits the success of his company's project so far to a "very strong business case." In addition, "we aggressively managed the schedule tightly, with daily conference calls, and paid attention to details and sizing," he says. By "sizing," Benson is referring to the process of estimating how large the individual activities are from the perspective of time and cost to complete.
Benson (who held the premerger title of vice president and CFO of NewTel Enterprises) personally put an enormous amount of effort into the ERP preparation stages. Working with a close-knit group of senior executives from each telephone company, he spent five months combing through business processes and working out the business case. Benson and his colleagues spoke to all the prospective users of the system, dissecting their jobs to compare what a process was costing in terms of labor and materials, then estimating the savings that might be realized using PeopleSoft.
"We had very tangible metrics," Benson recalls. For example, managers for purchasing and inventory studied how current processes would be changed under PeopleSoft and calculated the benefits of those changes, in terms of reduced inventory, lower costs through electronic purchase orders, and so on. Similarly, the accounts payable manager evaluated the volume of transactions handled under the old system and the associated discounts achieved with the vendor payment cycle, then projected the savings gained through automating parts of the cycle.
"We looked at every single business process that we would need to change," sums up Benson. "We identified all the potential cost savings to the lowest level we could possibly identify."
To make sure the estimated savings stuck, the group of executives reworked departmental budgets. "After we applied reasonableness checks to the estimates, we removed the amounts identified by the users from their budgets and replaced this with the estimates for the new system," says Benson. "Since compensation was affected by meeting those budgets," he adds, "our people were very committed to achieving what they said they could."
Having assigned executive-level "sponsors" to work with project managers for each area of business, Aliant began working "in bite-size chunks," beginning with the payroll system. Throughout the project, no changes to the PeopleSoft system were made. "We wanted to be able to simply drop in upgrades to the system without having to make modifications," explains Benson. "We estimated this would save us $3 million annually in information-system support costs."
But the "no-modification" plan demanded a $1 million training investment, as users were asked to forget everything they knew about their jobs and learn to deal with a completely new system. "Getting people to 'think outside the box' and learn to do their work differently was by far the most difficult task," says Benson.
Tightly managed as it was, the Aliant project nevertheless incurred unexpected costs. If problems with Process Number Two threatened to hold up Processes Four, Five, and Six, then Aliant bit the bullet and brought in consultants from Deloitte & Touche, PeopleSoft, and others. "[Consulting] cost us $500,000, but less than it would have had we delayed" completion of the project, says CFO Benson. And since other areas of the project were completed rapidly, he adds, "the [total project] costs evened out."
Back to Square One
What should a CFO do if the company's ERP project looks like it's about to become a white elephant? In such cases, the first step, other than calling in the consultants, is to build--or rebuild--the business case with clearly defined goals and terms.
"The CFO needs to be the person making the investment decision," says Meta's Grumberg. "He must insist that technology managers develop a language that can be understood. It's not appropriate to talk about 'bandwidth' or 'throughput.' What does that mean in dollars and cents, both in costs and benefits?" The business plan, he adds, should scrupulously weigh the investment in IT assets against a realistic set of risks, possible detours, and short- and long-term benefits. And everyone, whether in the executive suite or on the loading dock, must understand his or her role in the project.
In the end, ERP has relatively little to do with software, and it's about more than money. It's really about business management--the effort it takes to find all those tangible, intangible, and strategic benefits for which the software merely forms the infrastructure. It's also about building cohesive organizations and putting together better road maps that show potholes and detours. If nothing else, ERP projects offer the CFO the learning experience of a lifetime.
----------------------------------------------- --------------------------------- Trouble Spots
Establishing a solid, well-communicated ROI framework can help close what might be called the expectations gap. According to a recent study of 62 Fortune 500 companies conducted by Benchmarking Partners Inc. for Deloitte Consulting LLC, large variances exist between corporate expectations of ERP and what the systems actually deliver. For instance, most companies deploying ERP systems said they were not able to lower personnel, inventories, or maintenance costs as much as they had hoped. On the other hand, respondents noted better- than-expected results in overall productivity and in order-management cycle time, as well as in procurement, on-time delivery, and the ability to close financial cycles.
John Hunter, CFO of Decurion Corp., a privately owned movie-theater and real-estate development company in Los Angeles, thinks the gap may be due to lofty expectations set by software vendors, not management. "I have a general sense that CFOs too often succumb to the pressure of vendors that tell us what we could or should do, rather than doing what's right for our companies," he says.
The evil twin of expectation management is cost management. Hunter, for one, found that misaligned expectations result in inflated costs. Since 1992, Decurion has been running its general ledger, modeling, accounts payable, payroll, asset and property management, and time-accounting systems under J.D. Edwards's World ERP system (the company is currently installing the human resources, purchase order, and job-cost modules). Hunter says the original goal of the project was to automate and improve accounting processes while lowering inventory and labor costs. But, he concedes, getting there took more money and adjustments than anticipated. Consulting, programming, additional hardware, and training costs added $350,000 to the original investment of $1.2 million.
"We doubled costs for a solid year as we were implementing the system," recalls Hunter. "It was difficult for our staff to become proficient. We spent money over time customizing things."
Nor did benefits appear immediately. "We had a negative NPV for a year after installing the system," says Hunter. Such an experience isn't unusual--in fact, most companies take years after the software is rolled out to see any real benefits from their ERP systems, claims Debra Hofman, a research director with Benchmarking Partners.
"A quarter of the people we've spoken with say their key performance indicators took a nosedive after the 'go-live,'" she says. The reason? It takes a long time for companies to adapt to a new way of doing things. "The biggest issue is managing the change," says Hofman. In Benchmarking Partners's survey, the overwhelming majority of respondents listed change management, internal staff adequacy, and training as the biggest obstacles to achieving benefits from ERP.
Today, Decurion's ERP system is yielding substantial benefits. The financial operations staff has shrunk from 61 to 27; books are closed in 3 days rather than 30; and theater managers are able to better forecast staffing and inventory needs. But in hindsight, Hunter says, costs could have been better controlled had his company better understood that ERP isn't just about software. "We originally viewed this as [just] an accounting task, but it wasn't," he says. "The whole business needed to be involved. We missed out by not including people from functional areas in the decision process and setup."