UK card fraud losses on UK-issued cards reached £572.6 million in 2024 (UK Finance), and the great majority, £399.6 million or 69.8%, was card-not-present fraud: stolen card details used online or by phone. The real story is migration: as chip-and-PIN locked down the physical card, fraud moved online, even as the loss rate stayed tiny at 6p for every £100 spent.
Total card fraud losses, the dominant online category, and the loss rate that puts the pound figure in proportion.
Chip-and-PIN made the physical card hard to clone, so fraud followed the spending online. Card-not-present is now the dominant category by a wide margin.
| Fraud type | Loss (2024) | Share of card fraud |
|---|---|---|
| Card-not-present (remote purchase) | £399.6m | 69.8% |
| Other (lost/stolen, counterfeit, ID theft and more) | £173.0m | 30.2% |
| Total card fraud (UK-issued) | £572.6m | 100% |
At a physical terminal, chip-and-PIN and contactless verification make a stolen or fake card hard to use. Online, the fraudster only needs the card number, expiry and security code, which is why remote purchase fraud became the main battleground. Strong Customer Authentication, the two-step check now standard at UK online checkout, is designed to close that gap.
If you sell online, card fraud is your fraud problem far more than it is a high-street retailer's. The losses concentrate where the card is not physically present, which is exactly the channel e-commerce depends on. The cost of defending it, through authentication and chargebacks, is part of the card-acceptance picture on the UK card processing statistics hub.
Half a billion pounds sounds alarming, and for the victims it is. As a share of the money put through cards, though, the loss rate is very low.
UK Finance reports unauthorised card fraud at 6.0 pence per £100 of card spending in 2024, a rate of 0.06%. On a £1,000 of card turnover, that is 60p of fraud; on £100,000, it is £60. The pound total is large because UK card spending runs to trillions a year, not because any individual transaction is especially likely to be fraudulent.
That does not make the total trivial. It means the right way to read card-fraud statistics is alongside the spending they sit on, not in isolation. A rising pound figure can still mean a flat or falling rate if spending is growing faster.
For online sellers, the practical questions are authentication, chargebacks and the balance between blocking fraud and blocking genuine customers.
The two biggest UK fraud figures measure different crimes. Adding them together, or treating them as interchangeable, double-counts and misleads.
Card fraud here is unauthorised: someone uses your card details without your knowledge. APP fraud is authorised: you are tricked into sending the money yourself, usually by bank transfer rather than card. They are tracked separately by UK Finance, have different victims and protections, and should not be summed into a single "fraud" number.
For the authorised-push-payment side, including scam losses and the mandatory reimbursement regime, see UK APP fraud statistics and fraud reimbursement statistics.
Card fraud losses, the card-not-present split and the loss rate all come from UK Finance's Annual Fraud Report, the industry's authoritative fraud dataset.
| Source | Publisher | Period covered | Type | Last checked |
|---|---|---|---|---|
| Annual Fraud Report | UK Finance | 2024 (calendar year) | Industry body | 29 Apr 2026 |
The £572.6 million total and the £399.6 million card-not-present figure come directly from UK Finance's Annual Fraud Report (pages 22 and 25), the standard industry source.
The 6.0p-per-£100 loss rate is reported in the same report (page 28). We present it next to the pound total so the figure is read in proportion to spending.
The "other card fraud" figure is derived as the total minus the card-not-present figure, and is marked as derived.
All headline figures map to the UK Finance Annual Fraud Report. Card fraud (unauthorised) is kept distinct from APP fraud (authorised), the loss rate is shown beside the total, and the derived "other" figure is labelled. Last full review: 29 Apr 2026.