Facewatch facial recognition UK shoppers are now part of a routine retail reality, and the numbers behind it are larger than many people assumed. According to a report from Tech Times, the company fired 297,433 biometric alerts in six months of 2026. It then called on the Home Office to regulate an industry it dominates.

The report also says shoppers have no opt-out, and that the company's 99.98% accuracy claim excludes the human-review stage, where all three documented wrongful cases arose. This post sets out what is known, what is disputed, and what ordinary shoppers can do.

What 297,433 alerts mean for ordinary shoppers

An alert is generated when a camera matches a face entering a store against a watchlist of people a retailer has flagged. Across six months of 2026, the report counts 297,433 of them. That figure describes how often the system fired, not how many people were wrongly stopped, and it is worth keeping those two things separate.

What the number does show is scale. Live facial recognition in shops is no longer a pilot or an edge case. It is a standing feature of how some UK retailers manage theft, and shoppers may be scanned simply by walking through the door, whether or not they ever appear on a list.

The suggested concern here is less about any single alert and more about normalisation. When a system operates at this volume, errors that are rare in percentage terms can still affect real people. Retailers including Home Bargains have deployed the technology, as covered in our guide to your retail privacy rights around Facewatch scans.

Why shoppers have no opt-out

The report states that shoppers have no opt-out. In practical terms, the choice is to enter the store or not. Signage may tell you that facial recognition is in use, but a notice is not the same as a choice, and most people will not turn around at the entrance of a shop they need.

This matters because biometric data is different from most other personal data. A face cannot be changed like a password. Systems that process it continuously, for everyone who enters, shift the default from "you are anonymous unless suspected" to "you are checked unless you leave."

The retailer, rather than the shopper, decides who goes on a watchlist. The report highlights that this places a great deal of weight on internal review processes, which brings us to the accuracy question.

What the 99.98% accuracy claim leaves out

Facewatch advertises 99.98% accuracy. According to the report, that figure excludes the human-review stage, and that stage is where all three documented wrongful cases occurred.

That distinction is important. A matching algorithm may perform well on its own, yet the overall process also includes a person confirming or rejecting the match, then deciding what to do. If mistakes enter at that stage, a headline accuracy rate for the software alone does not tell a shopper how safe the full system is for them.

Two cautions apply. First, three documented cases is a small number, and we should not read it as a measure of how often errors happen. Second, documented cases are only those that came to light, and a wrongly flagged shopper may not know why they were challenged or how to complain. The useful takeaway is to ask what an accuracy figure actually measures, and who checks the result before someone is approached or banned.

What UK law and the Home Office consultation could change

Facewatch has asked the Home Office to regulate the industry. On its face, that is a request for clearer rules. The report's framing, though, is that rules written around the dominant provider could raise the bar for smaller rivals, locking them out while leaving the market leader well placed.

That tension is the heart of the debate. Good regulation could set standards on watchlist criteria, retention, review, and appeals. But who shapes the rules matters as much as what they say. Consumers, civil liberties groups, regulators, and competing vendors all have a stake, and a consultation process is where those views can be heard.

Until rules change, existing data protection law remains the main framework for how retailers handle biometric data, and shoppers keep rights such as asking what data is held about them.

What This Means For You

If you shop in the UK, you may be scanned without any practical way to decline. That does not mean you are on a watchlist, but it does mean your biometric data may be processed.

  • Look for facial recognition signage at store entrances.
  • If you are challenged or asked to leave, ask calmly why and who made the decision.
  • Request the data a retailer holds about you in writing.
  • If you believe you were wrongly flagged, ask for the decision to be reviewed and keep a record of what happened.
  • Pay attention to the Home Office consultation, since the rules set there will shape your rights.

Key takeaways

Facewatch facial recognition UK shoppers are being scanned at scale, with 297,433 alerts in six months, no opt-out, and an accuracy claim that does not cover the human-review step. The Home Office now has the chance to define the rules, and who influences them will matter.

For a practical walkthrough of how to ask what a retailer holds and how to challenge a wrongful flag, read our guide to Home Bargains and Facewatch scans.