Technophiles often tout RFID, or radio frequency identification, as the driver of the next revolution in the retail industry. But just when this revolution will sweep the industry -- and which vendors will benefit -- remains an open question.

"It's not 'if' anymore," Lyle Ginsburg, managing partner for technology innovation in Accenture's ( ACN) Global Products Operating Group, says of RFID, a World War II era technology that can help stores better track their inventory. But Ginsburg likens the current stage of RFID adoption to the start of a marathon: "It's going to be painful until you get your rhythm or routine down."

Understandably, many investors are struggling to find their stride as the potential RFID benefit is priced into names such as Zebra Technologies ( ZBRA), whose stock is trading at about 31 times forward earnings and 7 times sales, and Manhattan Associates ( MANH), a firm with a checkered past now trading at nearly 25 times forward earnings.

Meanwhile, RFID revenue remains relatively minuscule at the largest hardware companies such as Texas Instruments ( TXN) and Philips Electronics ( PHG ADR). In addition, a handful of interrelated hurdles -- standards, performance and price -- remain an issue for all RFID-related ventures.

Investing now in RFID is like asking how to make money on Y2K in 1994, warns Scott Lundstrum, a senior vice president and chief technology officer of AMR Research. "You can see the money out there, but you just can't get it yet."

Hype and Hope

The major reason for all the hype, of course, is a mandate from Wal-Mart ( WMT) requiring its top 100 suppliers to use RFID tags by January 2005 for merchandise destined for stores in Texas. Similar mandates have been issued by the Department of Defense, Target ( TGT), Albertson's ( ABS) and Tesco, the largest retailer in the U.K.

The hope is that RFID tags, which store product information much like a bar code, will help these entities track inventory better. The tag is a small microchip that can store more specific details about a product than a bar code and even be updated to follow a product's movement through the supply chain. Unlike bar codes, an RFID tag does not have to be directly in sight of a scanner to be read.

In addition to helping retailers with inventory, the technology is envisioned as a way to help detect counterfeits, track livestock and eventually prevent spoilage by storing temperature information. RFID is already the underlying technology in New York's EZ Pass and the Bay Area's FastPass toll-crossing systems.

Spending on RFID-related technology topped $1.9 billion worldwide last year and is expected to enjoy a 20% compound average annual growth rate through 2008, according to ABI Research, an Oyster Bay, N.Y.-based technology research think tank and consultancy. But for individual vendors, the RFID revenue won't be tangible for at least a couple of years.

Analysts are calling 2004 the year of the pilot as suppliers work to meet mandates from Wal-Mart and others. But the consensus today is that the tags must fall to the magic price of a nickel each to achieve widespread adoption.

Currently, the lowest published price is 19 cents a tag for 1 million tags, according to Accenture. AMR Research projects the number of tags shipped will not even cross the 200 million mark this year, but forecasts that number will skyrocket into the several billions by 2006.

RFID Picks and Pans

As is often the case with technology adoption, consultants are expected to be the initial winners in the early miles of the RFID marathon. Tony Ursillo, a hardware and software analyst with Loomis Sayles, points to Accenture as the obvious example, noting that RFID was a key technology component of Accenture's successful bid on the Department of Homeland Defense's massive $10 billion U.S.-Visit contract.

For now, RFID is not a big part of Accenture's business, but it's one of the company's fastest growing markets, Ginsburg said. "We believe over time it's going to grow and grow and grow," he said. "Thank you, Wal-Mart. It's certainly driving business our way."

Another oft-cited named is Zebra Technologies, which has parlayed its dominant position in bar code printing to become an early leader in the RFID market. Zebra is a favorite of Piper Jaffray senior analyst Brian Gesuale, who believes Zebra's RFID revenue should grow from far less than 5% of sales today to 10% by late 2005 and 15% by 2007.

However, investors may have already missed Zebra's run. Its shares have soared to more than $80, up 64% from their 52-week low.

Gesuale's retort: "It's a stock that is perennially expensive but continues to go higher."

In software, companies like Manhattan Associates, SAP ( SAP) and Retek ( RETK) are likely to have an advantage over RFID start-ups. That's because they already have a major place in retailers' operations, and such customers are more inclined to expand with their current vendors than add new ones, Ginsburg said. But AMR's Lundstrum believes software vendors won't see any meaningful opportunity from RFID until 2007 or 2008.

Notably, Fulcrum analyst Alan Weinfeld isn't including any revenue from RFID in his financial estimates for either Retek or Manhattan Associates in fiscal years 2004 and 2005.

Still, Weinfeld suggested Manhattan, which has positioned itself as an early leader with its "RFID in a box" product, is a sort of call option on the technology. Manhattan is working with 30 Wal-Mart suppliers moving toward complying with the retailer's RFID mandate, according to the company.

But Manhattan, whose major customers included beleaguered Kmart ( KMRT), has a mixed past, having lowered guidance for three consecutive quarters in 2003. More importantly, RFID revenue remains small -- only $500,000 from five small deals last quarter.

At only 10 times his 2005 earnings estimate when excluding cash, Retek's valuation offers better risk-reward than Manhattan, Weinfeld argued.

"No one has given Retek any credit for its warehouse management system, which is totally RFID-enabled," said Weinfeld, noting Target is among Retek's major customers. He believes Retek's shares should double within the next year. (Weinfeld's firm doesn't do investment banking.)

But Weinfeld's 2005 earnings estimate of 40 cents a share is significantly higher than the Street consensus of 26 cents a share gathered by Thomson First Call. And including cash, the stock now trades at 21 times the 2005 consensus estimate for earnings.

Another software company, VeriSign ( VRSN), sits in a unique position in the evolving RFID universe. As manager of the majority of Internet domain names, VeriSign routes roughly 10 billion Internet Web site look-ups per day. The company recently won a contract to manage a similar directory envisioned for RFID's electronic codes.

But again, this win won't immediately boost VeriSign's top line, in part because Wal-Mart initially is bypassing the company's central directory. Consequently, analysts aren't expecting VeriSign to begin to benefit from RFID until 2005 at the earliest.

By then, retailers should be closer to hitting their stride with RFID and the painful early miles may be behind them.