I’ve been tracking RFID for the twenty years I’ve been in the analyst business. Originally worked on at MIT in Boston, it was viewed as a panacea for inventory visibility and accuracy, both in the supply chain and in stores. Yes, I was always a doubting Thomas unless certain conditions could be met.

While there are some who think RFID is imperative in stores to keep track of inventory counts, there’s a very real reason why it hasn’t caught on in a broad-based way, particularly in the US. Overall, it has to do with the mechanics of physical inventories that are actually taken in stores, and the processes, procedures and systems required to ensure those counts are accurate.

Certainly, RFID makes it easier to find specific products. Several years ago, I went to Tyco’s retail sample store in Boca Raton (which was built in concert with Macy’s) and even way back then, an associate could scan and count 56 items a second using a hand-held scanner. I was impressed, and certainly, RFID can help make the invisible (one item in a sea of similar looking skus) visible. That saves workers a lot of time when fulfilling an order placed elsewhere. It can also save a customer time when looking, say, for the right size pair of jeans on a large set of shelving.

We know that the cost of physical inventories is the dirty little secret of retail, and if we could really do cycle counts and/or more frequent physical inventories in stores, the ROI is there both in cost avoidance and in sales uptick (how many “out of stocks” are really “I just can’t finds?”). 

There are those, including both partner Brian Kilcourse and me, who say that the Walmart mandate of 2005 set RFID adoption back at least a decade. The mandate was a solution looking for a problem. Few would deny that now. Further, if we assume that one value of carton-level RFID would be the “free read” item or carton-level RFID would provide to big box retailers when a carton was brought from the back room to the selling floor, share that with Consumer Products manufacturers and help trigger them to make more, rather than waiting for the demand signal from the Point of Sale.

Today, RFID and its “category,” the Internet of Things are becoming more commonly used, inside and outside of retail. We all use RFID technology when our cars whisk under readers that see the device in our “Easy Passes” (each state, and even county has its own name for the technology – long story short, no toll booths).

Yet the store remains essentially RFID-less for day-to-day operations.  Why?

The problem has never been the cost of tags. It has always been the sticky wicket of the processes, procedures and systems required to ensure those counts are accurate and the cost of RFID readers.

To get to an accurate physical inventory, or even quasi-accurate, which is more the reality of the situation, one needs cut-offs (establishing what should be considered counted, vs. in-transit, waiting to be received). One also needs to be able to count the entire store. It’s not like a warehouse, where product stays where you put it until a “move” directive is issues. 

In stores, the customer is the variable. She picks stuff up, and puts it down, wherever. And then there are end-caps. A retailer can decide to put an item both in its regular slot and also on an end-cap for promotion. Someone must count both.

Now, there is a truth – if a retailer decides to buy into cashierless checkout, there is likely no reason a full-store count cannot be implemented as well.  If.

Also, there is really no reason back rooms can’t be organized in such a way that, if nothing else, a store can count what’s in them without human intervention. I confess to have pondered this notion for a very long time. Nike has implemented something of a backroom locator system in stores. Boxes of shoes in back rooms will stay put.

Do I think we’ll see broad-based adoption of RFID and/or other IoT technologies across the body of stores any time soon? It would be great if we could/would, for sure. Will it happen? That will be a math problem retailers will have to sort through.  My absolute best advice is to gather up all the hard and soft costs of a physical inventory, and the cost of lost sales. Then one must factor in the cost of systems that ensure units and dollars (in other words, no “memo shrink”) are kept in synch and can consume the data generated.  And of course, in high volume stores, determine if the “pipe” sending information back to the home office is wide enough to accommodate bigger mountains of data.

I have been around the retail industry for a very long time. The quest to keep units and dollars in synch has been a long one and was only achieved about 15 years ago.  We could say “Well, now they’re both wrong,” but from a financial perspective, they’re probably quite close. It matters more to customers. and NO auditor is going to sign off on booking a cycle count in an aisle of a store.  None.

I really do look forward to the day when these problems can be solved. We are writing a supply chain report that talks, again, about lack of visibility, and the challenges of inaccurate inventory. We should be getting to a point where readers of all sorts are affordable. Can specialty stores lead the way, vs. the big box retailer?  It’s entirely possible. But then, I’ve always thought that way.