You've heard the war stories of enterprise resource planning system rollouts -- they're notoriously difficult to pull off. But supply chain management system projects may be even harder, given that they must unite not only the often-warring fiefdoms within a corporation but also the thousands of suppliers outside it.
Maybe that's why supply chain management is still in its infancy and no one has yet accomplished a fully integrated supply chain system, according to an in-depth study of 254 companies by researchers Stanley E. Fawcett and Gregory M. Magnan for the Center for Advanced Purchasing Studies (CAPS) in Tempe, Ariz. "Nobody is managing the entire supply chain from suppliers' supplier to customers' customer," the 2001 study concludes.
The potential payoff can be huge. A study by Peter J. Metz, executive director of the Massachusetts Institute of Technology Center for eBusiness, found that some companies achieve impressive bottom-line benefits, such as a 50 percent inventory reduction or a 40 percent increase in on-time deliveries.
But then there are the disasters, such as Beaverton, Ore.-based Nike Inc.'s announcement in March that installation problems involving new supply chain applications were largely to blame for an embarrassing profit shortfall.
Most supply chain projects probably fall somewhere in between. "What usually happens with these projects is that the momentum wanes as people have to work hard and make it happen," says Robert E. Sabath, director of the supply chain practice at Deloitte Consulting LLC in Chicago.
Computerworld talked with Sabath and two other supply chain consultants Jeff Kavanaugh, vice president of supply chain management at Inforte Corp. in Chicago, and Brent Habig, CEO of Tigris Corp. in New York to identify the major barriers to supply chain success.
1 Departmental warfare. Or, at least, passive resistance. Supply chain projects involve multiple departments, such as purchasing, planning, manufacturing, distribution and IT. At worst, this can mean turf wars. Even at best, different departments may support the overall goal but have conflicting priorities.
"More than hostility, you see people dragging their feet," Habig says. One classic conflict: Marketing folks want to have plenty of inventory of a blockbuster product to make sure they don't run out, whereas the goal of the inventory manager is minimal inventory. Balancing those interests is a "significant political challenge," Habig says.
2 Weak leadership. Because multiple departments are involved, supply chain projects need a strong, senior-level executive who can "be the tiebreaker," Kavanaugh says. "It requires leadership, because there will be challenges that come up. There needs to be a person of strong character who makes it happen . . . because there will be a thousand reasons why something might not work."
Habig agrees, suggesting that the top sponsor needs to be at the executive vice president level or that perhaps a senior steering committee should act as the project's board of directors.
3 ROI is hard to measure. This is one reason it's difficult to get strong executive sponsorship. If the return on investment is simply "greater customer loyalty," that's a hard sell with the board of directors, given the competing demands for capital investment.
The problem is that "the costs are not hard to quantify. The benefits are soft, but there are a lot of hard dollars being spent," Kavanaugh says. It's especially tough for public companies worried about quarterly earnings reports because it can take years to see the bottom-line benefits.
4 Funding problems. A company may decide to spend $30 million on a supply chain system, but if it doesn't make its quarterly numbers, the quick fix is to put the expensive project on the back burner. When that happens, IT project members move on, momentum is lost and consultants leave (and you don't get the same ones back later), Sabath says.
Kavanaugh says many companies are moving to a new strategy of "dynamic budgeting," meaning they have a list of priority projects to fund. "As long as you're working on the top couple (of items), you're fine," he says. "But if you're No. 10 and revenues go down, maybe they don't fund past No. 7." So sustaining funding for the life of a supply chain project that may take four to six years, is challenging, to say the least.
5 Scope creep. Imagine a typical project design meeting, in which each department chimes in to suggest great features the supply chain system should have. "Many times, we see projects over time get overcomplicated" with feature add-ons, Habig says. "By the end of that, you end up with a bloated project that is so extensive that it will take forever to accomplish, or if you do accomplish it, you don't get a solid return on the investment" because it's so expensive, he says.
Someone (see No. 2) has to make the tough calls, keep the project focused on the core business benefits and "throw some things out of scope," Habig says.
A related problem is trying to optimize the supply chain for every customer and product. "But you can't do that for every customer, every product, every transaction. If you do, the very best customers get shortchanged. That will bring down the new system," Sabath says. It's much better to make sure the supply chain works flawlessly for the top five customers and the most profitable products, Sabath emphasizes. In other words, some customers and some products are simply more important than others.
6 Legacy systems aren't integrated. Maybe the new supply chain system works fine but there are islands of incompatible legacy systems that weren't integrated. "Sometimes they can be integrated at a transaction level but not at the overall management level for common reporting and analysis," Habig says. "You need to create a layer on top of these systems that aren't talking that pulls them together so that you have visibility over the full picture." That would allow you to extract data for a consolidated report, for example.
Just think: What gives a company its competitive edge is probably housed in those customized legacy systems, Kavanaugh says. "The fact that the new packaged applications don't always link back to them is a problem," he says. "A new supply chain system won't be valuable if you don't bring along those competitive advantages, so you need to integrate those legacy systems."
7 Getting small businesses on board. True supply chain integration beyond the first tier of big suppliers is rare, the CAPS study concludes. "The task of managing beyond the first tier is handed off to the first tier with only minimal measurement and follow-up," the study says. Part of the problem is that small businesses many of which are quite happy relying on faxes, e-mail and spreadsheets often lack the resources to automate up to the standards of the big companies.
"The way you solve that is for the channel master to use software that can import and export data from spreadsheets," Kavanaugh says. "In other words, allow someone to exist at the spreadsheet level. And the good news is that, with a properly formatted spreadsheet, you can collaborate with that just fine."
8 Data quality problems. The CAPS study found that the No. 1 requirement for a successful supply chain integration is this: "Establish information systems capable of sharing real-time, accurate and relevant information." But lots more attention is lavished on the IT the "piping" than on the "quality of the data and making sure it's the right data for the decisions that need to happen," Habig says.
"You need highly structured, standardized bills of material, catalogs, part substitution (a taxonomy that recognizes substitutable parts) and standards between different companies and different industries," Kavanaugh says. "Content is a huge, if not glamorous, part of supply chain problems."
9 Sabotage by well-meaning employees. Sometimes your best employees the ones really trying to please the customer are the ones who foul up the new supply chain system. For example, if salesmen who need 100 of an item request 110 just to make sure they'll have enough, then the company has a false sense of demand, and inventories will swell.
The key to preventing this is to change the way employee performance is measured, Kavanaugh says. "If the salesperson is measured on how accurate they are in predicting sales, not just in how many they sell, those problems will start to go away," he says.
10 Old habits die hard. Remember that supply chain work was getting done before the software came along. So the biggest competition for supply chain software is the status quo: e-mail, faxes, the phone and spreadsheets.
Furthermore, a certain cynicism is creeping in. "Many materials managers view supply chain management as another management fad," the CAPS study says. "They believe that the term is beginning to mean 'everything and nothing' at the same time." wTips for SuccessOK, so we've identified the barriers to successful supply chain projects. But how do you overcome them?
-- Gain support from senior executives and departmental managers.
-- Ensure that the company is ready for the software project, with adequate resources, re-engineered processes and high-quality data.
-- Recognize that the supply chain software market is immature and more than one vendor may be required.
-- Define the requirements before evaluating various software products.
-- Test software products for scalability before purchasing.
-- Simplify the supply chain network by reducing the number of suppliers, then form close relationships with carefully selected suppliers. Not all supply chain partners are equal.
-- Start with a pilot project involving one or two suppliers and limited transactional data sharing.
-- Consider outside hosting services for business-to-business collaboration systems.
-- Encourage open and honest information-sharing.