How Do You Prevent POS Override Abuse in Retail?

By awareretailgroup.com Editorial· Edited by Tom Walsh· Published 2026-10-26

A POS override lets a cashier or manager change a scanned price or apply a discount, and it becomes abuse when the key rings an unauthorized markdown, such as a discount for a friend. When clearance markdowns also run through the override key, the register can record an abusive override the same way it records a legitimate clearance sale. Stopping it starts by pricing routine discounts into the register so the override key is reserved for real exceptions, then reviewing whoever approves those exceptions.

Why routine markdowns hide the overrides that matter

Famous Footwear’s loss prevention team says “one of the easiest routes for a cashier to take to steal is to under-ring an expensive item by doing a price override” (LP Magazine). The case study was first published in 2005 and updated in January 2016. The chain found that “about 40 percent of our markdowns are done via price overrides” (LP Magazine), including many clearance markdowns. Once that many markdowns run through a single key, the pattern “greatly obscures the real out-of-the-ordinary transactions that exception reporting looks for” (LP Magazine). An exception report built to catch an unusual override can lose a theft override among legitimate clearance sales when both use the same keystroke. For sorting a training problem from a policy or fraud problem, see training needs analysis for retail.

Remove the override key before adding approval layers

Famous Footwear’s answer was to change its POS procedures so routine markdowns no longer needed the override key, which the team called “a significant win for us” (LP Magazine). Of the unacceptable override uses the team tracked, it says price overrides were “our single biggest ROI area of impact from unacceptable uses” (LP Magazine). Taking routine markdowns off the key comes first because any approval or exception step added later then looks at real exceptions instead of clearance noise. Discounts carry the same exposure. Parker Griebel of the Wisconsin Organized Retail Crime Association says employees sometimes “override systems for friends or family, or they’ll use their own accounts to abuse digital discounts” (LP Magazine), which is why “internal controls and monitoring are so important” (LP Magazine) to limit the opening in the first place.

Why manager approval needs its own review

The Association of Certified Fraud Examiners analyzed 1,921 occupational fraud cases investigated between January 2022 and September 2023 for its 2024 Report to the Nations (ACFE). These cases come from organizations of all kinds, not only retailers. It found that a lack of internal controls was the primary weakness in 32 percent of those frauds, followed by an override of existing internal controls at 19 percent, a broader category than the POS override key (ACFE). A lack of internal controls was the top contributor at every level, but frauds by mid-level managers and employees were more often tied to a missing management-review step, while frauds by owners and executives leaned more toward weak tone at the top (ACFE). Management review was present in 72 percent of cases, and where it was, the median loss was $100,000, against $250,000 where it was absent, 60 percent lower (ACFE). ACFE has since published Occupational Fraud 2026, the 14th edition covering 2,402 cases; the figures here come from the 2024 edition (ACFE). An approval nobody reviews can let the approver’s own overrides pass unchecked. The override matrix below should name who audits the approver as well as the cashier. For the broader psychology behind why an employee reaches for that key, see why good employees steal.

What if the manager approving overrides is the one abusing them?

The ACFE finding above ties frauds by mid-level managers and employees to a missing management-review step, so a manager’s own overrides need a reviewer outside the store. Route every override a manager initiates to the next level up, and flag a manager who keeps approving overrides for the same cashier, as the matrix below does. Loss Prevention Media’s guidance on store closings recommends limiting select override authorizations to senior managers, since refunds and price changes should be minimal while a store closes (LP Magazine). A store in normal trading could pilot the same rule once routine markdowns are priced into the system and its override volume is already low.

What an exception report should flag

Griebel says point-of-sale systems already reject coupons that don’t match an item or exceed certain limits, and that “exception reporting helps flag unusual patterns like high-dollar redemptions or repeated overrides” (LP Magazine). A report that only totals override dollars misses the pattern that matters most: who keeps showing up on both sides of the transaction. It should also let a manager compare override rates across cashiers and stores against a peer average, since a cashier running double the store average is a different problem than a store running double the district average.

Does point-of-sale monitoring change behavior?

One large before-and-after study of point-of-sale monitoring comes from outside retail. Lamar Pierce of Washington University in St. Louis and co-authors studied 392 restaurants in 39 states before and after they installed point-of-sale monitoring software (ScienceDaily). Revenue per restaurant rose an average of $2,982 a week, about 7 percent, and theft dropped 22 percent after installation (ScienceDaily). The result suggests that monitoring register data can lower theft in a frontline service job. Catching theft after the fact is rare: ECR Retail Loss estimates, without naming the study behind the figure, that less than 2 percent of dishonest employees are apprehended (ECR Retail Loss).

What NRF’s retail security survey does and doesn’t measure

NRF’s 2023 National Retail Security Survey, conducted with the Loss Prevention Research Council and covering fiscal 2022, reported internal employee theft at 29 percent of shrink and external theft, including ORC-related events, at 36 percent (NRF). The average loss per internal theft investigation was $2,180 (NRF). Its list of internal theft methods names merchandise theft, refund fraud, cash and deposit theft, and sweethearting as the top four (NRF). The 2023 edition is also the most recent of its kind: Popular Information reported that the NRF did not release its annual survey after that edition and instead released a differently built report, The Impact of Retail Theft & Violence 2024 (Popular Information). A store that tightens override approvals is managing a mechanism this survey’s top four methods do not name separately, so cite it for the size of the internal theft problem rather than as proof that override controls move the number. For what to put in front of associates, see what employee theft awareness training should cover and how to reduce employee theft at the register.

Build your own override approval matrix

Every dollar limit, approver role and review interval in the matrix below is a pilot hypothesis, so treat it as a starting template rather than a benchmark. A store piloting senior-manager-only approval would move the shift-lead rows in the matrix up one level. Fill in each blank with current override volume and policy, then test it against a month of real transactions before treating any number as fixed. Expect the limits to differ by format: a c-store running one cashier on a shift may need remote or after-the-fact approval where a grocery store with a dedicated service desk can require a second person on the spot.

Questions to ask your POS vendor about exception reports

Before the matrix above means anything, confirm the exception report behind it can actually answer these questions.

• Does every override record the initiator’s ID, the approver’s ID, the reason code, and both the original and final price as separate fields?
• Can the report flag an override where the approver and the initiator are the same person, or where the same pair repeats?
• Can legitimate markdowns, promotions, and gift-with-purchase pricing be loaded into the system so they never touch the override key at all?
• Can override rates be compared across cashiers, shifts, and stores against a peer average, not just a flat dollar ceiling?
• Can an override be linked to the loyalty or customer ID on the same transaction?
• Can approval thresholds be set separately by role and by override type, with every threshold change logged?

Any question the vendor answers no to belongs in the next contract or configuration request.