Over the past two years, some of the biggest names in retail have been quietly taking self checkout lanes back out. Walmart pulled the machines from a handful of stores. Target capped its self checkout at ten items or fewer. Dollar General removed self checkout from its 300 worst stores for loss and put a staffed hand back on the register in roughly 9,000 more. None of these companies is short on technology or budget. They walked the lanes back anyway, and the reason they gave, over and over, was shrink.
The self checkout kiosk was supposed to be a win on every count: fewer labor hours, shorter lines, a customer who serves themselves. For a lot of stores it delivered on the first two and quietly gave the savings back through the third. The question is not whether self checkout is good or bad. It is what a kiosk can and cannot see, and what fills the gap when the cashier is no longer standing there.
What the kiosk was built to see
A self checkout station has one job at the moment of sale: record what the customer scans. It is very good at that job. It reads a barcode, adds a line item, takes the payment. What it was never built to do is judge whether the scan matches what is actually in the basket. It does not know that a customer waved a steak over the glass without a beep. It does not know that two items went into the bag and one went around the scanner. The kiosk records the story the customer tells it, and it has no way to check that story against the shelf.
At a staffed register, a person closes that gap without thinking about it. A cashier notices the item that did not beep, the bag that looks heavier than the receipt, the small hesitation. That noticing is never written down anywhere, which is exactly why it is easy to cut. When a store moves transactions to self checkout, it does not only move the scanning. It removes the quiet human check that used to sit on top of the scanning, and nothing takes its place.
What the research actually found
This is not a hunch. Adrian Beck, an emeritus professor at the University of Leicester, measured the loss across 93 retailers in 25 countries for the ECR Retail Loss group. The finding was blunt. For every 1 percent of sales a store pushes through fixed self checkout, its unknown loss rises by about 1 basis point. Put half your sales through the kiosks and that is roughly half a point of extra loss you cannot see and cannot name. Two thirds of the retailers in the study said the problem was getting worse, not better.
The wider theft picture is moving the same way. The National Retail Federation reported that the average number of shoplifting incidents rose 93 percent in 2023 compared with 2019, with the dollar loss up 90 percent. The machines that removed the human check arrived just as theft climbed, and that is why the big chains reversed course. As Target's chief executive put it, this is not one company's problem, it is a retail problem. The convenience industry's own trade press tracked the same pullback, with Dollar General reporting its shrink climbing more than 100 basis points in a single year.
What a connected loop sees that the kiosk cannot
Pulling the lanes back out is one answer. It trades the labor savings back for the human check, and for some stores that math works. But it treats the kiosk as the problem, when the real problem is narrower. The kiosk is blind to everything above the barcode, and the barcode is only one system in the store. There is also a camera pointed at that same lane, and it sees exactly what the kiosk misses.
That camera, on its own, is no better than the kiosk. It records the steak crossing the glass in full detail, but has no idea whether the steak was ever rung, because it cannot read the transaction. The kiosk has the transaction and not the floor. The camera has the floor and not the transaction. The loss lives in the seam between them, which is precisely why a store can own both and still not catch it.
The thing worth building is not a better kiosk or a better camera. It is the join. ARGUS runs what we call the closed loop. It crosses the camera with the point of sale, continuously, on the lanes a store already has. It watches items physically move through the self checkout and watches the transaction record at the same instant, then flags the moments where the two do not agree. An item that crosses the scan zone with no line item is a mismatch. A basket that leaves heavier than the receipt is a mismatch. Nobody has to suspect anything first: the loop raises the flag on its own, attaches the clip, and ranks it by cost, turning a drive full of footage into a short list worth reviewing.
Why this matters more at a convenience store
For a convenience store or a gas station, the stakes are not the same as they are for a national chain. They are higher. A grocery chain can absorb half a point of extra loss across enormous volume. A single site running on thin margins does not have that cushion. A point of shrink at a store that keeps a few cents on the dollar is not a rounding error. It is the difference between a good month and a flat one. That quiet loss is real money leaving a specific building every shift.
The point of measuring loss at the lane is not to catch one thief. It is to make the invisible number visible, so the loss shows up on a daily Revenue Integrity Score instead of arriving as a surprise at the next stocktake. A store that can see the gap can decide what to do about it, whether that is coaching, staffing, a lane change, or simply knowing which register to watch.
Self checkout is not going away, and it does not need to. What it needs is the check that used to stand beside it. ARGUS puts that check back without putting the labor back, by letting the camera and the register do together what neither can do alone. We are in private beta with convenience, gas station, and grocery operators. If self checkout loss is the part of shrink you cannot get your arms around, talk to us or write to business@useargus.co.
Sources
- ECR Retail Loss: Adrian Beck's global study on self checkout loss, measuring how self checkout adoption tracks with unknown loss across 93 retailers in 25 countries.
- National Retail Federation: The Impact of Retail Theft and Violence 2024, which reports a 93 percent rise in the average number of shoplifting incidents since 2019.
- NPR: Reporting on major retailers reversing course and scaling back self checkout with item limits and store removals.
- NACS: The convenience industry trade association's coverage of Walmart, Target, and Dollar General scaling back self checkout as shrink rose.