They say good things come to those who wait. Ed Matthews had better hope so. His company has already spent $250K on RFID equipment with no prospect of payback for another year—or two—or three.

When Matthews accepted the information systems director's post at Pacific Cycle, he undoubtedly assumed he'd spend his days untangling knotty systems integration problems. What he had no way of knowing was that he'd soon be spending most of his time immersed in the murky world of radio-frequency identification. Or that the skill he'd find most valuable would not be his wide-ranging systems expertise but a hitherto undiscovered ability to "sell" management on pilot programs that would burn through hundreds of thousands of dollars with only the remotest prospects of returns.

It all started in June 2003, when Wal-Mart issued its now infamous mandate requiring its top 100 suppliers to be RFID-ready by January 2005. Pacific Cycle, which supplies the mega-retailer with Schwinn, Mongoose, Murray and other brands of bicycles, was one of those top 100 suppliers. Though no one had any clue how the company would meet this demand, at least three things were clear: They would need to come up with a compliance plan quickly. They would need time and resources to test whatever system was chosen. And they would need funds.

It fell to Matthews, the company's director of information systems, to go before the company's CFO to petition for funds to cover startup costs for an RFID pilot. As if begging for thousands of unbudgeted dollars wasn't bad enough, things only got tenser when the CFO asked about the expected payback time. Hemming and hawing, Matthews had to admit that it would be two, three,maybe four years. He also felt it was only fair to warn him that until Pacific Cycle figured out the best location for the RFID tags, it would be throwing away tags—worth an estimated 50 or 60 cents apiece—by the thousands.

In the end, of course, he got the OK—what vendor wants to risk losing Wal-Mart as a customer? But a few months later he was back in the CFO's office. The testing team had run up against a stumbling block and it was a big one. He was going to need more money.

The team had known from the start that reading accuracy would be a problem—Pacific Cycle ships mainly metal products, and metal is known to interfere with radio waves and compromise reading accuracy. Sure enough, during the first goround of tests, read accuracy rates were only in the 70-percent range. Matthews and his staff tinkered with tag and reader locations, but nothing worked. Finally, they decided to try tags from a different manufacturer—Matrics, a tag maker now owned by Symbol Technologies. Trouble was, Pacific Cycle had already sunk $100,000 or so into the original tags and hardware. Now Matthews had to go back to the CFO to get approval to purchase not only new tags, but new hardware as well—and with no guarantees that this would work either.

The tide turns
It was here that the team finally got its first break. They switched over to the new tags and equipment, and read rates immediately soared to near 95 percent, where they currently stand. (Matthews believes that as bigger players enter the tag market and the technology matures, read rates will inch even higher.)

Flush with success, Pacific Cycle's RFID pilot team went on to meet its first objective, shipping four pallets of bikes with RFID tags to Wal-Mart's Dallas DC in September, nearly four months ahead of schedule. "We're proud to be one of the first Wal-Mart suppliers to demonstrate the ability to meet the retailer's requirements," says Matthews. He admits, however, that questions about what information Wal-Mart will exchange with its suppliers still need to be worked out. For example, the first data Pacific Cycle received back from the retailer's DC did not include initial reads for bikes like the 24- and 26-inch models that are too large to be moved by conveyor. Matthews says that Wal-Mart is aware of the problem, and he expects a solution soon.

Still, this progress has come at a price. In late 2004, Matthews estimated that Pacific Cycle's RFID expenditures would reach $250,000 for equipment alone by year's end. Then there are the labor costs.When a shipment arrives at Pacific Cycle's 800,000-square-foot DC in Olney, Ill., workers have to identify the boxes of bikes headed for Wal-Mart, open them, and affix an RFID tag (usually to the owner's manual), and then reseal them. Why can't the workers simply slap a tag on the box? Matthews explains that won't work because the bicycles are taken out of the boxes for assembly once they arrive at Wal-Mart, and Pacific Cycle would lose its tracking ability at that point.

Despite Wal-Mart's assurances that its suppliers will benefit from installing RFID systems (increased supply chain visibility and a reduction in stockouts are the benefits most frequently mentioned), Matthews has no illusions about seeing a return on investment anytime soon. "There is not a clear return on investment out there at this point in time," he reports. "Right now, I don't think [we'll see] a full ROI until mid 2006 or 2007."

High hopes
Given Matthews' experiences, no one would blame him if he felt he had been taken for a ride. Yet he remains surprisingly optimistic. He rationalizes that the technology is still in its infancy—still on training wheels, if you will. And he remains convinced that when it matures—that is, when it's finally ready for the Tour de France—the returns will be well worth the wait.

In fact, Matthews is already thinking ahead to the next step. Though he has no timetable yet, he's already planning for the day when Pacific Cycle begins tagging not just pallets and cases, but all of its individual items. The ability to track bikes from the DC to Wal-Mart's back room and eventually to the retail floor will revolutionize the business, he says. Pacific Cycle will finally have reliable data on exactly which items are out on the retail floor and which lines need replenishing, allowing it to act on that information without delay. "If we see a lot of product in the back room at a particular store," Matthews explains, "we can send someone over to find out what the problem is and fix it."

And he can't help thinking about the labor savings the company will realize once he's able to push the RFID tagging process downstream to the manufacturing plants in China, relieving DC workers of the time-consuming work. That capability is still at least two years away, however. Right now, two obstacles stand in the way: First, China has not yet set standards that will allow RFID tags to be encoded in that country. Second, despite ramped-up RFID demands from customers like Wal-Mart and Target, Pacific Cycle's RFID volume is nowhere near high enough to justify the cost of shifting the tagging process to overseas plants. While he waits for the retailers to catch up, Matthews says, "we're pretty much doing a slap and ship because it's just not cost effective to do otherwise."

Just what exactly is Matthews hoping to achieve with RFID? Lower labor costs, a reduction in inventory and higher shipment accuracy, for starters. Once the company's operations are fully RFID-enabled, DC workers will no longer have to spend long hours filling out paper forms to receive, put away, pick and ship goods. Transactions will be instantly updated on Pacific Cycle's ERP system, eliminating the delays associated with manual entry. Access to location information on every item in its DC will allow the company to cut safety stocks and reduce back-orders. And the near-perfect accuracy provided by RFID technology should reduce time spent searching for product and virtually eliminate charge-backs from retailers.

But a lot of things have to happen first. Pacific Cycle can't expect to see any kind of return on its investment until it begins using RFID in its internal operations.And that won't happen until it's able to tag the majority of incoming products, says Matthews. That alone could take years. "We won't be able to justify that until tag prices drop and we come up with an inexpensive way to integrate the technology with our back-end SAP system," he says. "We also need our retailers and suppliers to fully utilize this technology so that we can leverage it over the entire supply chain and not just make this a retail or internal technology."

These are formidable challenges, to be sure, but Matthews is convinced Pacific Cycle will overcome them. The company has already solved tougher problems, he says, and besides, it has no choice. "RFID," says Matthews, "is clearly a technology that is here to stay."