Traceable Quality
Talk to us
Menu

Insight

Where RFID earns its money

Count one rack of 400 cartons with a scanner and with a reader, then the three questions that send you back to the barcode.

Count a rack of 400 cartons.

With a barcode, somebody walks up to each carton, finds the label and points a scanner at it. Call it three seconds a carton if the label faces out and nothing’s in the way. That’s twenty minutes for the rack, and the top shelf needs a ladder.

With RFID you walk past it. The reader hears every tag that gets enough power, through the cardboard, hundreds of them a second. The walking is what takes the time. About a minute.

Twenty minutes with a scanner, about a minute with a reader. Then the three questions that send you back to the barcode.

Every RFID pitch deck has a version of that comparison, and it’s true. It’s also the easy half of the decision.

The case for RFID

It’s really three cases.

A scanner can’t reach the label. The carton is shrink-wrapped on a pallet, or turned the wrong way, or two metres up. A barcode needs a clear line from the scanner to the label. A UHF tag needs radio, and cardboard is close to invisible to radio.

There are too many to scan one at a time. A reader settles which tag talks when, so a few hundred tags in the field all get heard in one pass. Readers built on Impinj’s E710 chip run anywhere from about 50 tags a second in their longest-range mode to more than 1,000 in their fastest, and Zebra quotes 1,300+ a second for its RFD40 handheld. Nobody’s shelves move that fast. You’re limited by your feet.

Nobody’s there to scan. This is the one that changes the job instead of speeding it up. Put a reader on the dock door and the pallet gets counted on its way through. Nobody scanned anything. The count stopped being a task.

So when is the barcode the better buy?

When the item is cheap.

A plain paper UHF label lists at ₹5 to ₹10 in India, more if you want it printed and encoded in small runs, and several times that for on-metal tags. Take ₹10. On a ₹300 item that’s noise. On a ₹30 item it’s a third of the price, on every unit you ever ship, and the label doesn’t get cheaper because the item did. A barcode costs fifty paise of label stock.

When someone’s already scanning.

A packer scans the carton to close the order. A cashier scans it at the till. That scan is already paid for, and it does a job the tag can’t skip, because the order doesn’t close itself. Tag that carton and you’ve added a cost and saved nothing.

When you need to know exactly which one.

A read tells you the tag is somewhere in the antenna’s field. That field can be a few metres across. For a stock count, “it’s in this zone” is all you wanted. For a serialised part going into a specific slot on a specific job, it isn’t. A scan says: this one, in my hand. RFID can get closer with more antennas and more software, and by then you’re paying a lot to imitate a barcode.

The numbers, and where they come from

The three seconds and the one minute are assumptions. Here’s what they rest on.

The best independent count I know of is still the University of Arkansas pilot at Bloomingdale’s in 2009. Staff counted 209 items an hour with barcodes and 4,767 an hour with RFID. 209 an hour is seventeen seconds an item, and almost none of that is the scan. It’s finding the item, turning it over and finding the label. So three seconds a carton is generous to the barcode, because it leaves out the finding.

The RFID side of that pilot was a shop floor of folded clothes, which is slower to sweep than a rack of cartons. So the film’s minute for 400 cartons is optimistic for a store and fair for a rack. If your stock is the folded-clothes kind, put 75 seconds per hundred into the demo instead of 15 and watch what it does to the answer.

And the time saved on the count isn’t where most of the money is. The bigger saving is finding the cartons that were put back in the wrong place, and that depends entirely on how often it happens to you. I left it out of the demo on purpose.

Try it on your own count

Barcode or RFID? asks for the count you actually do: how many items, how often, what they’re worth, whether a scanner can reach the labels, whether someone scans them anyway, and whether you need the exact unit. It gives a verdict, the reasons behind it, and the hours each way. Every assumption above is a field you can change.

For the longer comparison with QR and BLE, see RFID vs barcode vs QR vs BLE. For how a tag with no battery talks at all, there’s Wake the tag.

Change the tag price first, and use a quote, not a list price. Four hundred cartons in a minute is a lovely thing to watch. It’s also ₹4,000 of labels on one rack.

Next step

Tell us what you are trying to trace.

Describe the flow: what arrives, what happens to it and what leaves. We will tell you what it takes to capture it.