Getting Narrow With RFID Is The Only Way It Will Work

This week's Wall Street Journal article on Dole's use of RFID in the produce supply chain is yet another example of RFID being applied for specific use-cases, rather than in a blanket effort to "improve visibility."

In response to the E. coli outbreaks from last fall, Dole and other producers are starting to tag produce from field to store in an effort to introduce trace-ability into the supply chain and make it easier for health officials to narrow in on sources of outbreaks. Coupled with Wal-Mart's increasing focus on in-store applications, the success of recent CPG pilots around in-store promotional displays, and successful pilots by Best Buy and Marks & Spencer tagging high-value items in the store, RFID use is fast moving from general applications to specific uses within the retail supply chain.

Why get specific? At an aggregate level, it's easy to see the value that RFID can bring to the supply chain. If you know where every case is from the moment it rolls off the production line until it hits the trash compactor at the store, then theoretically you can solve almost any supply chain snag or at least have all the information you need to recover when something goes awry.

The problem is, that doesn't work in reality: Visibility alone is not enough to build a business case.

Yes, you can make assumptions about how many supply chain problems would be avoided if you had perfect information (presumably supplied by RFID) and that could potentially be enough to get some numbers on a spreadsheet to look good enough if tag prices get low enough.

However, all "visibility" gets you is enough data to start asking questions that identify issues. If you don't know the right questions to ask, then you won't identify issues, and you won't get any value out of visibility. That's why these specific use-cases � trace-ability, promotional displays, counterfeiting, and high value item tracking � are getting better traction. There's a problem to solve; people already know what questions they need answered.

The root cause of supply chain issues is not "lack of visibility." When Kraft, Unilever, and Proctor & Gamble started testing RFID-tagged in-aisle promotional displays, they found a host of root causes � everything from a true lack of visibility problem (stock clerks not being able to find it in the back room of the store), to basic operations and training issues (like when the stock clerk pulled the product out of the display, stocked it on the shelf, and trashed the display itself).

P&G can impact the first problem � by shipping the display much closer to the floor date � but they have no control over the latter problem, other than to complain to the retailer when it happens. Visibility reveals problems to be solved � it doesn't solve them. And it takes time to make the fixes needed to actually solve the root cause of the problem.

When early leaders in RFID looked to blanket a supply chain with tags and readers, the objectives focused on open standards and interoperability. The whole point, after all, is to share data with trading partners to enable "visibility." But this new, use-case driven RFID is different.

Use-cases drive unique requirements, not standards. Dole wants to trace boxes of lettuce from field to store. Best Buy wants to inventory a tightly packed shelf of video games in ten seconds instead of four hours. A department store retailer wants a consumer to be able to hold a shoe up to a kiosk and get size availability information. These are completely different uses of the same technology, requiring different methods of tagging and reading, different frequencies and different types of tags.

In a way, it means a step back for RFID, as companies tinker with closed-loop systems that solve their unique problems first, and then potentially later consider wider implications. While it means more pain later on, loosening the emphasis on developing standards could actually speed adoption today.

The blanket mindset no longer applies. In the wide application of RFID, every case has to be tagged and every trading partner has to be reading in order for visibility to work. It doesn't really help to trace one box through the supply chain when it could take one of potentially hundreds of routes from source to shelf.

But limited applications do help specific use-cases. CPG tagging of promotional displays proved that nicely: they didn't have to wire up every case and every store in order to get benefits from solving the root causes they identified.

A process change that changes behavior at a wired store �like shipping promotional displays closer to the floor date so there is less chance of it being lost in the back � can have just as much impact at an unwired store. Removing the blanket requirement from an RFID implementation can turn a business case from a risky maybe to a slam dunk.