Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
E-commerce payment fraud is the full set of ways an online retailer loses money at or after checkout, from stolen cards and account takeover to promotion abuse, refund fraud, and false disputes.

What Is E-Commerce Payment Fraud?
Ask a retailer what fraud costs them and the answer is usually a chargeback figure. That number is the smallest part of the picture. A merchant also loses the goods, the shipping, the fee, the staff time, and, far larger than all of it, the revenue from genuine customers whose orders were declined because they resembled a fraudster. Any serious fraud programme measures all five.
Rules are fast to write and fast to evade. Models are harder to game and harder to explain. Most effective programmes run both: deterministic rules for the things you are certain about, such as blocked addresses and velocity ceilings, and a scoring model for the grey middle, with manual review reserved for genuinely ambiguous cases rather than for everything the rules flagged.
The signals that carry the most weight are the ones attackers cannot cheaply change: device fingerprint and its history with your site, email address age and reputation, the relationship between billing and delivery address, behaviour on the page, basket composition and how it was assembled, and the customer’s prior order history. Card details alone say almost nothing, because they are the part the fraudster bought.
Feedback is what turns a model into a system. Chargebacks arrive weeks later, so unless disputes, manual review decisions, and confirmed fraud are fed back into the model, it degrades steadily while appearing to work. Measure the approval rate, the fraud rate, the manual review rate, and the dispute ratio together, and treat a change in any one of them as something to explain rather than celebrate.
False declines are the under-managed half of the problem. A declined genuine customer is a lost order, a lost lifetime relationship, and sometimes a complaint on social media, and in most retail categories the value of wrongly rejected orders exceeds the fraud losses avoided. Set an explicit tolerance, review rejected orders as carefully as accepted frauds, and give the fraud team a target that includes approval rate rather than losses alone.
A UK merchant operates under strong customer authentication, fully enforced for e-commerce card payments since 14 March 2022, which has pushed losses towards first-party disputes and account takeover. The Payment Systems Regulator’s reimbursement requirement for authorised push payment scams, in force since 7 October 2024, has also changed how bank transfer fraud is handled, although it does not cover card purchases.
Turkey looks different. Authentication on domestic card-not-present transactions is close to universal, so stolen card use is comparatively rare. Instalment payment, a standard feature of Turkish card products, creates its own abuse patterns, and cash on delivery shifts risk into the logistics chain described under delivery fraud. A fraud model trained on UK data will perform poorly in Turkey, and the reverse is equally true.
Underneath all of this sits the platform itself. Price manipulation, coupon stacking, and unverified payment callbacks are application flaws, not fraud patterns, and no scoring model will catch them. At Privia Security, we test checkout and order flows against both abuse and exploitation through our Penetration Test Services.
A three-party scheme. The fraudster advertises goods cheaply on a marketplace, collects payment from a genuine buyer, then orders the item from a legitimate retailer using stolen card details and ships it to that buyer. The buyer receives real goods and never complains, the retailer loses the item and the chargeback, and the fraudster is left with clean money. Delivery address checks do not help, because the address is genuine.
In most retail categories, yes, by value. Every wrongly declined order loses the sale, the margin, and often the customer relationship, and the volume of declined genuine orders typically exceeds confirmed fraud by a wide margin. Set an explicit tolerance, review declines with the same rigour as confirmed fraud, and include approval rate in how the fraud team is measured.
Treat the sensitive action rather than the login as the control point. Require re-authentication to add or change a delivery address, to use a saved card with a new address, or to change contact details, and notify the customer out of band when any of these change. Combine that with detection of credential stuffing at the login endpoint and with monitoring for unusual ordering behaviour on established accounts.
Buy the baseline, build the parts specific to your business. Commercial platforms bring consortium data on devices, email addresses, and known fraud patterns that no single merchant can reproduce. Your own team knows which basket combinations, promotions, and product categories are abused in your business, and those rules belong in-house, reviewed regularly as attackers adapt to them.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.