PRIMAL FEAR

When the buzz about collaborative commerce began a few years back, it sounded like a winner: Retailers and vendors would use the Web to forecast better how much of what products they needed. Stores would move more merchandise, suppliers could produce just enough, and everyone would make more money.

Five years later, retailers and suppliers are still experimenting with collaboration, many through a process now called CPFR — collaborative planning, forecasting, and replenishment. And many of them are still reporting positive results. “But even in those companies that have tried collaborative planning … oftentimes it’s one part of the organization that will try it, in some cases with some successes, but it’s not totally widespread,” says Hau L. Lee, a professor of operations at Stanford University. And that has a few people beginning to wonder: Why, if CPFR is so great, isn’t everybody doing it?

It’s not that the magic doesn’t work

Dee Biggs, director of customer logistics at Welch’s, the Concord, MA-based grape juice maker, says that following the nine-step CPFR process with a major retailer helped Welch’s decrease forecasting errors from 40% to a weighted average of 8% within one year. And Welch’s is apparently not alone in seeing such stunning results. “Every initiative that we’re aware of has increased sales — sales have gone up anywhere from 3% to 20%,” says Joe Andraski, a consultant at OMI International in Atlanta and vice chair of the CPFR committee at the Voluntary Interindustry Commerce Standards Association.

It’s not that the technology is too tricky

Supply chain experts say that unlike some systems, there’s not a whole lot to go wrong with the collaborative forecasting software. “The technology has never been the issue, or at least in the last five years has not been the issue,” says John E. Rutherford, an industry veteran and adjunct professor of supply chain management at Golden Gate University in San Francisco. The Web-based applications for CPFR primarily compare the sales forecasts of vendor and retailer, alert managers from each team when items aren’t moving according to plan, and provide a bulletin board to help the partners iron out differences.

It’s not that suppliers don’t like it

Some suppliers who have tried it, who might be justifiably suspicious pursuing an initiative pushed by Wal-Mart and other big retailers, report that joint forecasting through the CPFR model has actually helped them reduce the number of unwelcome surprises. In the past, for example, extra demand from store promotions would occasionally come to suppliers without warning — and without sufficient inventory to back them up. One collaboration expert for a vitamin supplier says that without CPFR, it’s common for vendors to find out only after the fact that a retailer ran a promotion. “In the past it was almost like, ‘Holy mackerel, did you see these numbers? What did they do?’” recalls Terry Moody, collaborative forecast planner for Pharmavite of Northridge, CA.

So what’s holding collaborative partnerships back? People. Sometimes people in the supplier’s company. Sometimes people in the retailer’s. Unfortunately, supply chain experts say, CPFR is more about communication than technology. And as anyone who has spent any time in the human race can tell you, clear communication between individuals is difficult to achieve. Between large and complex businesses, each with its own vocabulary, its own systems, and its own perspective, the challenge is even greater.

Talking the talk

While lack of trust between partners is frequently cited as a reason for rocky CPFR pilot projects, collaboration experts say that the more typical challenge is making sure that the information being shared is based on common measurements and a common vocabulary. Some of this is easier said than done: Stanford University’s Lee says that he’s seen situations, for example, in which two companies agreed to share forecasts on a monthly basis, only to discover that one company was measuring the month from the first to the first, while the other’s month ran from the 15th to the 15th.

Often, too, even if the techies can synchronize the data between two partners, the people running a CPFR partnership don’t actually know everything that’s going on in their company — a condition that supply chain experts say is all too common. Steve Banker, service director for supply chain management at ARC Research in Dedham, MA, says this is because most companies don’t have enterprise-wide forecasts yet. “I wouldn’t recommend trying CPFR until you’ve already achieved a one-enterprise forecast,” he suggests. “If in your enterprise there is a forecast for marketing, a forecast for manufacturing, and a forecast for distribution, and you’re all working on different forecasts, that’s the low-hanging fruit — you [should] tackle that problem first.”

Another reason for the seemingly endless succession of CPFR pilots, what Lee calls “toy” programs, may be that the total impact of CPFR on the supply chain has not yet been well-enough measured to induce companies to make a greater investment in the process. Without such measurements, many executives remain unconvinced. In a recent survey of 150 Fortune 1000 executives, consulting firm Accenture found that a majority, 54%, agreed that developing collaborative relationships was important. However, some were unsure about the ultimate goal: 26% percent of the executives surveyed cited “lack of a clear value proposition” as the biggest barrier to collaboration. That survey dealt with collaborative strategies in general, but such uncertainty seems to be true of CPFR in particular as well. In a March report on CPFR in the grocery business, written by KJR Consulting for the Grocery Manufacturers of America and Syncra Systems, a collaborative software company, researchers said that many grocery manufacturing executives said they were enthusiastic about CPFR — but unsure about what it actually was.

But CEO inattention isn’t the only way to kill change, of course. Other pilots fail through more democratic means. “There are many pilots that fail because management wanted to do it but the people who actually live with the day-to-day interactions resisted it,” Lee says. “It’s not enough to have the bottom level, but it’s also not enough to have the top level.”

Don’t call the taxidermist yet

Some experts say that the relatively slow adoption of CPFR may also be because it’s a bit less of a miracle elixir than the hype suggested. “I think the vision of doing CPFR across all SKUs and across all partners was probably unrealistic to begin with,” says Banker.

“I think maybe initially we were thinking, maybe there’s 20% of the SKUs, 20% of the suppliers you don’t want to do this with,” he adds. “I think now it’s shifted the other way, where we’re saying, maybe there’s 80% of the SKUs and 80% of the suppliers we don’t want to do this with.”

Why? Some suppliers just aren’t sophisticated enough to hold up their end, experts say. Or their relationships with the retailer are not solid enough to withstand the openness between buyer and seller that’s required. Still other firms may be supplying products that are too unusual or too slow-moving to warrant such strategic scrutiny.

But the biggest reason for the relatively slow march toward the land of collaborative replenishment may be that the promoters left one minor detail out of the brochure: CPFR might not require a lot of money up front, or a huge investment in technology, but it does require a fair amount of work — meetings, conference calls, software training, and more conference calls.

“That was the part that people didn’t talk about early on,” says Hank Steerman of Sears, who has led four CPFR initiatives in the past two years. On average, Steerman says, it takes about 90 days for his team to sort out all the details involved in creating a CPFR relationship with a single vendor. “The value that we’re seeing does live up to some of the hype, but it’s not something that you basically just say, OK, we’re going to do this starting Monday, plug it in and begin to get results.”

At the same time, it’s probably premature to start working on the diorama for CPFR in the hall of extinct business ideas. For one thing, mass merchandisers are still reportedly committed to the process, and this typically causes the hearts and minds of suppliers to follow. “When we talk about collaboration, collaboration usually means coercion,” Banker explains. “Usually, where ‘collaboration’ begins is where you have a Kmart or a Wal-Mart demand that you do it or else. Supply chains where you have an 800-pound gorilla being served by a band of chimps, and the 800-pound gorilla says jump, are the supply chains where you see ‘collaboration.’”

For another, people keep learning more about CPFR through pilot projects, and veterans evangelize its benefits. “When you get into the trading partner relationship, you start understanding where you’ve been screwing each other up or causing people to do things that didn’t make money,” says Jim Uchneat of Benchmarking Partners in Cambridge, MA.

Finally, another inter-industry initiative is under way that should help make the CPFR process easier. The Uniform Code Council, the folks who brought us the bar code, are working now through their subsidiary, UCCnet, to create more common data standards. Uchneat says that in addition to many other benefits, greater data synchronization between departments and between companies should help simplify the CPFR process.

“It’s doable,” says Biggs of Welch’s, reflecting on working with CPFR. “It’s a better idea in an industry that really doesn’t forecast demand very well. This is an opportunity to do a better job of that — to start to communicate.”

You say you want an evolution

Andraski doesn’t sound disappointed by the piecemeal progress CPFR has made so far. “None of this supply chain stuff was started by anybody who looked at it and said, ‘I’m going to have a full-blown supply chain management system,’” he says. “This has all been evolutionary.”

It’s true that change sometimes does take time. After all, it’s taken more than 50 years for the bar code to emerge from the mind of a bright young graduate student in 1948 and into widespread circulation on the nation’s grocery shelves. Even now, scanners are found in only 60% of U.S. grocery stores, although the bar code’s cost-saving benefits have been touted since 1970.

“Start small, keep it simple, and have minimal expectations,” Andraski says. “Because when you start looking at it from the standpoint of trying to put together a completely holistic collaborative program within your company, it gets overwhelming.”

Bennett Voyles is a business writer based in New York City. He can be reached by e-mail at [email protected].

Looking for the right match

Collaboration experts say that one of the big keys to success is taking some time to pick the right partner for a CPFR pilot. What should you look for? Someone you can trust. “A CPFR initiative can only really work between partners who trust each other. If you have a relationship [in which] the retailer and supplier distrust each other, and don’t want to share information, CPFR will never work,” says Hank Steerman of Sears. Someone fairly large. You’re probably going to be better off working with a reasonably large company, one that has both a sophisticated IT department and forecasting specialists, advises Steve Banker of ARC Research. Someone with a strategic product. Choosing a company that supplies a typical product is a good idea, experts say. You don’t want to set up a pilot project that is such a special case that you can’t apply what you learn to the rest of your business, says Hau L. Lee, a professor of operations at Stanford University.