The prospect of affordable tags has retailers drooling. If every item in a shop were tagged, an apparel retailer, say, could both improve customer service and combat top-line losses, which are typically 5 to 15 percent of sales. RFID technology could be used to locate mislaid products, to deter theft, and even to offer customers personalized sales pitches through displays mounted in dressing rooms. Ultimately, tags and readers could replace bar codes and checkout labor altogether.
Affordable RFID tags also have enormous implications for the supply chain: since all EPCs are unique, products tagged in factories could be tracked as they moved along the supply chain. The resulting visibility of inventory levels and flows of raw materials could result in large savings. We estimate that a retailer or consumer goods maker using RFID could cut total warehouse labor costs by nearly 3 percent, chiefly through more efficient receiving, shipping, and exception handling. More promising still are the potential effects of RFID on vendor-managed inventory systems. By exchanging the information gleaned from RFID readers over the Internet, a consumer goods maker could manage its own stock replenishment for key customers more efficiently, saving both parties 20 to 40 percent or more in inventory and out-of-stock costs.
But retailers and suppliers alike should cast a gimlet eye over the current state of the technology. The main value of RFID is that it eliminates the need to handle items individually—by allowing, say, distribution centers to receive mixed pallets of goods automatically. But if the tags themselves are not robust, reliable, and tamperproof, the savings evaporate. Tellingly, an October 2002 pilot by the Auto-ID Center found that RFID-tagged pallets failed 3 percent of the time even when double-tagged; only 78 percent of the individually tagged pallets were read accurately.
Companies should avoid fixating on the price of a tag lest they lose sight of the costly upgrades in enterprise-resource-planning (ERP) software that RFID technology requires. For relatively easy tasks, such as measuring inventory levels, simple add-ons might suffice. But tackling more complex applications, including tracking individual items throughout the supply chain, would require ERP upgrades that might cost tens or hundreds of millions of dollars for a large company. Server and network infrastructure would also need fortifying to handle the thousands of additional data transactions per product.
So the watchword, for both retailers and manufacturers of consumer products, is caution. For most retailers, long-term investments in RFID technology are still risky. They should be undertaken only by the small number of industry-dominating companies, such as Wal-Mart, that have the clout to influence suppliers as well as the deep pockets to weather the ups and downs inherent in using a nascent, albeit promising, technology. Wal-Mart’s recent announcements indicate that it plans to proceed slowly, concentrating first on pallet- and case-level tagging to reduce supply-chain-handling and inventory costs. [The McKinsey Quarterly]