First-Party Fraud Is Quietly Rewriting Your Loss Curves — and Most Programs Are Not Ready
Friendly fraud and dispute abuse have quietly become a top loss driver for issuers and merchants. Why traditional fraud stacks fall short — and what mature programs are doing differently.
When fraud teams talk about fraud, they usually mean third parties — stolen cards, account takeovers, synthetic identities. That is the fraud their tools, dashboards, and staffing models were built around.
The fastest-growing line on most issuer and merchant loss reports is not third-party fraud. It is the cardholder themselves disputing transactions they actually made — what the industry calls first-party fraud, or friendly fraud.
It is no longer a rounding error. For many programs, it is a top-three loss driver — and the operating model used to fight it was designed for a different problem.
Why It Is Growing Now
A few things have compounded:
- Digital commerce has trained consumers to expect frictionless dispute experiences, and those experiences have made disputing easier than returning
- Subscription, marketplace, and digital-goods volume — all categories where dispute claims are hardest to refute — has grown faster than card volume overall
- A small but visible "chargeback as a hack" subculture circulates instructions on social platforms, lowering the perceived stigma and effort
- Issuer dispute teams, sized for a different fraud mix, often find it cheaper to credit the customer than to fight a low-value claim
Why Traditional Fraud Stacks Don't Catch It
Conventional fraud stacks score the transaction. They are very good at distinguishing a stranger using a stolen card from the legitimate cardholder. They are essentially blind to a legitimate cardholder who buys something they intend to dispute later.
By the time the signal exists — a chargeback claim — the transaction is days or weeks old, the merchant has shipped, and the dispute lands in a queue sized for a smaller, mostly third-party problem.
The Operational Gap
Most issuer dispute operations have two structural problems:
- Staffing levels and SLAs were set against a chargeback mix that no longer exists
- Compelling-evidence rule changes from the networks (CE 3.0 and the equivalents) have shifted both responsibility and the win-rate ceiling, but few teams have re-trained representment workflows around the new evidence standards
What Mature Programs Are Doing
The programs making real progress have stopped treating first-party fraud as a fraud problem and started treating it as a customer, product, and ops problem at the same time. Common moves include:
- Building cardholder-level dispute behavior signals — repeat disputers, dispute velocity, and patterns relative to spend — into both authorization and dispute decisioning
- Investing in the merchant evidence pipeline so representment is built around the data the networks now require, not the data the dispute team happens to have
- Re-pricing or re-segmenting the categories most exposed (digital goods, subscriptions, marketplaces) rather than absorbing the loss across the whole portfolio
- Closing the loop with merchant partners, where the merchant has data the issuer cannot see
The Honest Question
Most programs do not need a new fraud platform to address first-party fraud. They need an honest look at where the losses are actually coming from, and a willingness to redesign the operating model around the answer. The programs that do that work over the next 12 months will look very different on a P&L from the ones that don't.