Where the Tag Goes On: Source, Warehouse, or Store
The most consequential RFID decision is not on the shop floor. It is where the tag gets applied, and how you know it was done right.
Most RFID stories start on the shop floor. A handheld sweeps a rail, the screen fills with a stock count, and someone says the accuracy number out loud like it settles the matter. It does not. The decision that actually sets your accuracy and your cost was made long before the tag ever reached the store. It was made the moment someone chose where the tag goes on.
There are three honest answers: at the source, in the warehouse, or in the store. They are not interchangeable. Each one moves cost, control, and risk to a different place in your operation, and across a fleet of hundreds of stores those differences do not add up. They multiply.
Why placement is the real decision
A single tag costs a few cents. That number is so small it stops people thinking. But the tag is the least of it. What you are really choosing is who applies it, who encodes it, who checks it, and who is accountable when it is wrong. Get that right once and it scales cleanly across every store you own. Get it wrong and you are paying for the mistake in every location, every day, in stock numbers you cannot trust.
So the question is not “should we use RFID.” It is “where does the tag go on, and how do I know it was done right.” Hold onto that second half. It runs through everything below.
Source tagging: cheapest, best, and the furthest from your control
Source tagging means the RFID label is applied and encoded at the point of manufacture. The garment arrives at your distribution center already carrying a tag, usually a UHF EPC Gen2 inlay encoded with an SGTIN, the serialised item number that makes every piece individually unique.
On paper this is the winning option, and on a fleet it usually is. The cost of applying the tag disappears into a process the factory is already running. There is no extra handling step in your own operation. And because the tag goes on at the start of the journey, your store accuracy tends to be the best of the three, because the item has been readable and identifiable the whole way to the floor.
The catch is the one you would expect. You have outsourced the quality of your inventory data to your suppliers. If a factory tags the wrong item, encodes the serial incorrectly, or skips a percentage of the run, you do not find out at the source. You find out weeks later, downstream, when a store’s numbers drift and nobody can say why. The cheapest tag in the world is worthless if you cannot trust that it was applied correctly and that it is actually there.
Warehouse tagging: control bought with throughput
Tagging in the distribution center is the middle path. The goods arrive untagged, and you apply and encode the tags yourself before they ship to stores. You own the process, which means you can inspect it, measure it, and fix it without waiting on a supplier.
That control is real and it is worth a lot, especially while your supplier base is still maturing toward reliable source tagging. But it is not free. You have added a step to the busiest, most time-pressured part of your operation. Every unit now passes through a tagging and verification station, and that costs labor, floor space, and throughput in a building whose entire job is to move product out the door quickly. For a high-volume DC, that drag is not a rounding error.
Store tagging: full control, no scale
Tagging in the store gives you the most direct control of all and the least of everything else. Staff apply tags to items as they arrive or as they are put out. It works. It is also the most expensive option in labor, the most error-prone, and the one that scales worst across a fleet, because you are now running a small tagging operation in every single location, staffed by people whose actual job is selling.
There are moments when it is the only option: a pilot, a category your suppliers cannot yet tag, a one-off correction. As a standing model across hundreds of stores it throws away most of what made RFID worth doing, which was getting the supply chain to carry the load for you.
The real question: how do I know it is right?
This is the part the vendor demo skips, and it is the part that decides whether any of the above actually works.
The further upstream you tag, the less you see with your own eyes, so the verification has to be built in. Retailer RFID mandates have made this concrete: read rate expectations now sit at or above 99 percent in a realistic store environment, and a tagging non-compliance rate of even ten to fifteen percent is enough to wreck inventory accuracy and make every audit suspect. That is the line between a system you trust and a number you quietly stop believing.
So you build the checks where the goods change hands. The most important one is the inbound audit at the distribution center: as cartons arrive, RFID readers confirm that what was ordered is what was tagged, that the encoding matches the purchase order, and that nothing was missed. Tags that are wrong or absent get caught at the door, not on the shop floor six weeks later.
Around that sits a supplier program. You measure each supplier over time on encoding accuracy, placement consistency, and read rate, and you treat tagging quality as a compliance metric with consequences, the same way retailers use chargebacks to make the cost of bad tagging land on the party that caused it. The point is not to punish. The point is that accountability has to live somewhere specific, because data quality that is everyone’s responsibility is no one’s.
A short framework before you choose
Before you pick a placement model, answer these honestly:
- Volume. How many units, across how many stores? Higher volume pushes you upstream, toward source.
- Supplier maturity. Can your suppliers tag and encode reliably today, or only in a slide? If not yet, warehouse tagging buys you time without giving up control.
- DC capacity. Can your distribution centers absorb a tagging and verification step without choking throughput?
- Accuracy requirement. What does the business actually need on the floor, and what read rate does that imply?
- Verification capability. Can you check tag quality at the point goods change hands? If you cannot verify it, you cannot trust it, and placement is moot.
The bottom line
The cheapest, best-performing tag is the one applied furthest upstream, at the source. But that is only true if you can trust that it was done right, and trust at that distance does not come for free. It comes from inbound audits, supplier accountability, and the discipline to measure tagging quality as seriously as you measure the accuracy number it produces.
Where the tag goes on is the easy half of the decision. How you know it is right is the half that actually runs the store.
-Bart