Radio frequency identification has been considered an important emerging technology since June 11, 2003, when Wal-Mart CIO Linda Dillman said the retail giant would require its top 100 suppliers to put RFID tags on pallets and cases starting in January 2005. Six years later, Wal-Mart still doesn’t have RFID interrogators in every store, and few other retailers have jumped on the RFID bandwagon. Why?
There’s no simple answer. Many factors contribute to the speed at which technology adoption takes place, and some adoptions that seem to take place overnight really have been incubating for years. RFID is not as easy to deploy as, say, desktop computers, Internet technologies or enterprise resource planning (ERP) software. Computers require asset purchases and training, but simplify manual tasks, such as managing budgets and typing letters. Internet and Web technologies require a lot of cabling and coding, but deployments remain largely within the IT department’s control. Even ERP systems, which involve big, complex software installations, typically involve IT managing deployments and training.
|
 |
Click here to view a larger version of the chart.
RFID deployments are more complex, because they involve dealing with the physics of radio waves. For some products that are RF-friendly, such as apparel and paper products, this isn’t a problem. But for products that contain a lot of water—most meat and vegetables, for instance—or are packaged in or made of metal—foil-lined bags, food cans and so on—the RF issues can be challenging. The RFID industry has overcome them, to a large degree, but the perception that RFID won’t work around water and metal persists in the mainstream business press and in the minds of some businesspeople.
In addition to physics, there is the challenge of how to deal with the data collected. Because each tagged item is associated with a unique serial number, and tags can be read many times by one or more readers, companies must be able to filter the data and integrate it into their back-end systems. End users rank this as their biggest concern about RFID deployments, according to some surveys.
This challenge may not be as massive as is generally perceived. During the RFID in Fashion event hosted by RFID Journal and the American Apparel and Footwear Association in New York in August, Zander Livingston, director of RFID at American Apparel, said his company has set up in-store systems that use RFID to manage store inventory and replenishment; the data is transferred to existing back-end systems, where it’s treated as bar-code data. Simon Langford, director of EPC RFID strategies at Wal-Mart, said his company is taking essentially the same approach.
The bigger obstacle to RFID adoption is a lack of understanding about what data can be collected, and how it can be used to reengineer business processes and reduce costs or increase sales. Nowhere are the benefits clearer than in the retail apparel sector, where studies by the University of Arkansas’ RFID Research Center show that inventory accuracy is roughly 60 percent to 65 percent—meaning 35 percent of the time, stores have inventory they don’t know they have, or don’t have inventory they think they have.
American Apparel has improved inventory accuracy to 99 percent in its five RFID-enabled stores, according to Livingston. It spent about $50,000 per store for three fixed interrogators and three or four handheld devices, as well as software to manage inventory replenishment, and sales at these stores have risen, on average, 14.3 percent, Livingston says. But even American Apparel isn’t aggressively deploying the technology. The company is committed to RFID, Livingston said at RFID in Fashion 2009, but given the current economic downturn, it’s trying to decide whether to use existing capital to open new stores or to RFID-enable more stores. He’s currently doing a controlled study to quantify the benefits more precisely and will present the data to the board in the fall. After that, the company will decide how quickly to roll out the technology in 2010.
Wal-Mart knows it will achieve significant benefits by deploying RFID, but mindful that its suppliers are reluctant to absorb tag costs, the retailer has been trying to quantify the benefits for them. In 2007, it launched a trial in which it tagged all cases of air fresheners shipped to several RFID-enabled stores. Sales increased for all types of air fresheners, the study showed, because RFID improved inventory visibility and replenishment. In 2008, Wal-Mart extended the category tagging to more types of products. The retailer hasn’t publicized the results of the expanded trial, but several people whose companies were involved told RFID Journal the results were mixed: In some cases, the ROI for suppliers was clear, but in others it wasn’t. Wal-Mart says it’s still committed to RFID and is seeking to collaborate with suppliers to achieve mutual benefits, but it hasn’t spoken publicly about plans since its Sam’s Club division relaxed the deadline by which all suppliers must tag at the item level.
Some RFID solutions providers that have been involved in retail pilots say a major obstacle to adoption is cultural. “The retail industry traditionally has not spent a lot of money on technology, and it continues to believe that retailing is more an art than a science,” says one systems integrator who didn’t want to be named for fear of alienating clients. “Retailers don’t understand how RFID can improve their processes, and when you show them how much sales can be improved, they simply refuse to believe it.”
To be sure, RFID doesn’t deliver much benefit when it’s used to tag slow-moving, low-margin items. But for apparel, footwear and jewelry, the benefits can be huge. Stores can sell more by improving inventory accuracy—RFID lets you take inventory in 96 percent less time than bar codes do, according to the RFID Research Center—and replenish more effectively. RFID also enables managers to monitor how well store associates execute tasks, because each item is tracked as it’s sold and replenished. RFID Journal‘s Fashion Retail ROI Calculator shows that an average specialty apparel retailer could achieve a return on investment on each store system in seven months. It also could achieve an additional $13 million in sales across 400 stores simply by selling more items at or nearer to full price. (The ROI calculator is available for free at www.rfidjournal.com/calculator.)
Despite this, one retailer told RFID Journal that RFID is difficult to deploy in stores because, unlike pure software solutions, it requires IT teams to visit stores and deploy hardware. Tags have to be applied to items already in stores. Store managers have to be convinced to embrace the technology. Employees in stores have to be educated and trained on a new system. “It’s messy stuff,” the retailer said. “The IT departments don’t like to venture out into the stores. They like the controlled environment of the headquarters.”
Still, RFID is close to achieving a breakthrough in the apparel sector, at least. Bloomingdale’s, after an initial pilot, has decided to tag all items in one store. JC Penney has done a pilot, and Dillard’s has been doing extensive testing with the RFID Research Center.
In all likelihood, one major retailer will adopt RFID to track individual clothing items within the next 12 months, and others will follow. “No one wants to be first,” says the systems integrator who worked on several retail pilots. “In this economic environment, CEOs are very cautious, but once someone provides the benefits, others will follow.”