Only the finest fabrics and top designers are good enough for upscale couturier Escada (USA) Inc. But when it comes to an enterprisewide application suite, market leader SAP AG’s R/3 is far too lavish, according to Ed Beesley, technology chief at the exclusive retail chain.
The cost of R/3 isn’t at issue. Rather, the increasingly notorious implementation hassles and associated delays in reaping promised business process gains pushed Beesley toward a less-vogue vendor, Lawson Software, for his core client/server applications.
Like many IS managers at midsize companies (Escada’s U.S. revenues were $160 million in 1995), one of Beesley’s top priorities is to replace aging minicomputer systems with high-performing client/server systems quickly, on budget, and with limited resources. Naturally, he considered buying R/3, the world’s leading client/server application suite. But after contemporaries at larger, wealthier R/3 sites regaled him with horror stories about exploding costs and project delays, Beesley got cold feet. “If I make the commitment to spend $250,000 to implement a financial system, including all the software, consulting, and training, I don’t have another shot to come back and say, ‘I need $50,000 or $100,000 more,'” says Beesley.
Customers like Beesley add a new twist to the challenges of implementing R/3. IS managers in the so-called middle-market–companies with yearly revenues between $50 million and $750 million–typically have small staffs, so they need more outside help when implementing R/3. Yet they also have smaller budgets and slimmer margins for error than their well-endowed colleagues at Fortune 500 companies. As far as R/3 is concerned, those two needs are at odds. So some of these middle-market managers are questioning whether the risks just might outweigh the benefits of deploying R/3.
These misgivings have sobering implications for SAP. R/3 is the trendiest software to hit corporate America since Lotus Notes. Users love the integrated nature of the client/server applications (financial, manufacturing, distribution, and human resources), and SAP has a commanding lead in the U.S. business applications market.
Yet despite SAP’s well-documented success in Fortune 500 companies, R/3 isn’t playing as well in America’s huge middle-market, where there’s still a huge untapped market and healthy competition.
The primary turn-off for midsize IS shops is R/3’s complexity. The integrated software modules offer a rich array of options, but before installing any of the modules, users must configure thousands of tables even if they don’t need all the functionality.
This complexity helps lead to the kind of costly and lengthy implementation that a company running on tight margins can ill afford. Industry experts estimate that midsize companies spend on average three to five times the cost of the software and take nine months to get R/3 running. That’s less than the 10-times software costs and several years implementation deploying R/3 can take in a big, bureaucratic company. However, it’s much more than the standard model for the 1980s host applications, which was $1 of implementation costs for every dollar of software cost, analysts say.
Also, middle-market companies’ lean IS staffs necessitate a greater reliance on costly outside expertise. Because R/3’s the rage in the Fortune 500, consultants command $1,500 per day on average. And even at that price, quality can be a big issue. Andy Hafer, the technology guru at Hydro Agri North America, in Tampa, Fla., went through eight different consulting houses while trying to implement R/3. He found himself paying $200 an hour for novice advisers, whom he had to “weed out” because they weren’t qualified.
As a result, midsize companies try to make do with less. At KPMG-SETAC, a Boca Raton, Fla., company that helps medium and large firms implement R/3, Vice President Olivia Sadati notes that large companies use two to three R/3 consultants and some business process re-engineering experts to help implement one accounting module. In contrast, a midsize company generally can only afford one consultant per module.
As soon as internal staffs are trained, companies run the risk of losing them to the consulting firms and Fortune 500 companies that can pay big salaries. One IS director at a midsize R/3 user site wouldn’t talk on record for fear of exciting recruiters. “Every time my name appears in a SAP story, the phones start ringing off the hook with headhunter calls,” he says.
Despite all this, nobody sees the middle-market completely shunning R/3. Some midsize companies, especially those operating in multiple foreign currencies or planning rapid expansion, are embracing R/3, challenges and all. They want to guarantee that they won’t outgrow their business software. “It was a much bigger system than we needed today,” says Ron Munden, senior vice president of Accugraph Corp., in El Paso, Texas. However, the company is laying the groundwork to grow to $100 million quickly. “Switching accounting systems is painful, so we want to be able to use it for a long time,” he says.
To cope with the implementation process, some of these companies apply the industry axiom known as the 80/20 rule. The premise is that 80 percent of a system’s benefits come from deploying 20 percent of its capabilities. Conversely, to eke out the remaining 20 percent return requires 80 percent of the effort. By invoking the 80/20 rule, IS managers can get R/3 up and running faster than if they try to deploy full functionality right off the bat.
Hydro Agri managed to deploy four R/3 modules in eight months by running only the features that enabled basic business functions such as accounts receivable. Since it went live a year and three months ago, the company has been adding some of the nice things that weren’t overly necessary at first, says Hafer.
All these issues are likely to change in the future. With the Fortune 500 market licked, SAP knows it has to win the middle ground to retain its eye-popping growth. The company is conducting a study, nicknamed Heidelberg, to determine how to simplify R/3 and expand services for middle-market needs and budgets, according to Bryan Plug, president of SAP Canada and the executive responsible for SAP’s middle-market initiative.
One plan is to add a set of “super switches” to R/3, which would allow SAP implementers to “turn on” various functions to meet a particular customer’s needs. Another option under consideration is to build a value-added reseller channel that would sell software and services for a fixed price.
But none of this is easy to deliver, and it won’t be available for a while. SAP already ditched plans for “R/3 Lite,” which was a reduced price subset of R/3 functions. Meanwhile, the Heidelberg plans won’t be announced until next year.
Yet even if SAP trims R/3’s implementation costs and complexity, the software won’t necessarily be natty enough to suit the middle market. When Beesley looked at R/3’s order-processing module recently, he found it didn’t have the features he needs to support Escada’s multiple inventory classifications. “I’m not trying to rip them apart,” he says. “[SAP] is really hot. But I just don’t understand it.”