A lead sales associate at a Dollar General in Louisiana pulls a stack of fresh price labels from the printer and drops them in the trash. The register already shows the new price. The shelf still shows the old one, and the company has paid over six million dollars in fines over seven years to keep it that way.

The price on the shelf is the price you pay. Most people who have run a business have trusted that line without checking it. It sounds right. It is not.

The FTC, joined by twenty states, filed suit against Dollar General for charging more at the register than the price shown on shelves and tags. The complaint documents 29,265 consumer reports of the same problem filed between early 2023 and early 2026. That is not a run of bad luck at a few stores. That is a pattern with a structure behind it.

Paragraph 93 of the FTC complaint names the structure. Register prices update on their own. Shelf labels require a person to print them, walk to the aisle, and swap them by hand. Most price changes are increases. When no one swaps the label, the error runs one direction. Toward overcharge. Every time.

FTC staff ran undercover shopping trips at Dollar General stores in four cities. Between 8.5% and 22.4% of the items rang up at a price that did not match the shelf. The net overcharge per trip ran small: $3.25 in one city, $4.07 in the next, $2.60, $3.65. Small per receipt. Millions of receipts per year.

Marc Dann, former Attorney General of Ohio, put it in plain terms. He called it "a crime committed, 25 cents at a time."

The problem is not that the labels are wrong. The problem is that fixing the labels costs more than leaving them wrong.

Dollar General runs each store with as few staff as possible, often leaving only one person on the floor for hours at a time. A typical small grocery store employs more than twelve. The company schedules a five-hour weekly shift to swap out old labels. Most stores are too thin on staff to run it. The register updates itself. The shelf waits for a person who is not coming.

Seven settlements with state regulators in seven years. Over six million dollars in total fines. In the same stretch, Dollar General posted annual revenue north of forty-two billion dollars. Six million against forty-two billion is 0.014% of one year's sales.

That is not a penalty. That is a licensing fee.

There is a name for this in the literature. Lund and Sarin, writing in The Regulatory Review, call it just another cost of doing business. When the expected gain from the practice exceeds the expected penalty, a firm that runs on profit will absorb the fine and keep the practice in place. It is not a moral failing on the part of one company. It is math. And the math says: keep going.

Most people treat this as a customer service failure. It is a pricing architecture.

The system is built so errors tilt toward the seller, and the cost of correction runs higher than the cost of getting caught. That same structure does not stop at a retail shelf. It sits inside any vendor, supplier, or contractor whose cost of fixing an invoice runs higher than the cost of a complaint. Every quoted rate you have accepted without checking carries the same incentive. The gap is not personal. It is structural. And it runs in the seller's favor by default.

The receipt is the only contract. Once that is clear, three moves follow from it.

Pull every invoice line and check it against the quoted rate before payment clears. Not after. Not at the end of the month. Before the money moves. This takes more time than most operators want to spend, and that is the exact gap the pricing architecture counts on.

Log the gap between the quoted price and the invoiced price for each vendor over 90 days. One mismatch is a typo. Two is a pattern. The log turns a gut feeling into a data point you can act on.

When the same vendor drifts more than once in a quarter, renegotiate the rate or replace the vendor. Do not send a polite email asking for a credit. Renegotiate the whole rate. The drift was not an accident. It was a feature of how that vendor prices.

Running this for a quarter does something the old advice never did:

It shows which vendors price clean, every time, with no drift. It shows which vendors drift, and by how much per invoice cycle. It shows where your own margin is leaking without a line item to explain it. And it shows how much of that gap you were absorbing before you knew it had a name.

At the end of 90 days, ask three things.

→ What moved the number?

→ What looked like a fair price but left no trace in the margin?

→ What drift showed up more than once from the same source?

That is the difference between advice that sounds right and a system that proves itself.

The tags are in the trash in Louisiana. The register shows one number. The shelf shows another. No one is coming to fix the gap.

The only person who catches it is the one who checks the receipt. That has always been true. Now you know the math behind it.