Friendly fraud — the 65–75% problem no vendor talks about honestly

What friendly fraud actually is

Friendly fraud (also called first-party fraud, chargeback abuse) is when a genuine cardholder disputes a genuine transaction. The card wasn't stolen. The customer really made the purchase. Then they file a chargeback.

Common patterns:

• Customer forgot they made the purchase (subscriptions especially)
• Customer received goods but disputes anyway ("I never got it" when they did)
• Family fraud — spouse or teenager used the card without primary knowing
• Post-purchase remorse — customer doesn't want the item, disputes rather than returning
• Deliberate abuse — customer knows they can get goods free by disputing

All of these show up under standard fraud reason codes (Visa 10.4, MC 4837, Amex F29). Indistinguishable from actual stolen-card fraud to the network's system.

Why it happens

Friendly fraud is a symptom of the chargeback system's design. Card networks give cardholders a low-friction dispute process — file with your bank, get provisional refund, no proof required. Merchants have to prove the transaction was legitimate. Default is buyer-favorable.

Digital-goods merchants suffer most — there's no package to prove delivery of. Chargebacks911's 2025 Annual Report puts digital-goods friendly fraud rates 40% higher than physical goods.

Subscription businesses face a related problem. Cardholders forget they signed up, see a recurring charge months later, and file instead of remembering.

Why prevention tools don't catch friendly fraud

Traditional fraud tools look for transaction-risk signals — mismatched addresses, unusual devices, velocity spikes, high-risk countries. Friendly fraud has none. Customer is real. Device is theirs. Address matches. AVS full match. CVV correct.

3DS authentication can help by shifting liability — but adds friction that can drop legitimate conversions 5–15%. And 3DS doesn't stop the dispute; it changes who absorbs the loss.

Verified customer accounts and device fingerprinting are useful, but they're evidence for AFTER the dispute — not prevention.

Why representment does catch friendly fraud

Representment — fighting the chargeback with evidence — is where friendly fraud gets caught. Strongest evidence categories:

• Delivery to AVS-verified address: proves goods reached the cardholder
• Prior undisputed transactions: proves ongoing relationship (CE 3.0 territory)
• Post-purchase login history: proves cardholder accessed the account
• Support-ticket history: proves cardholder communicated with you

CE 3.0 explicitly targets friendly fraud. The rule mandates issuer acceptance when merchants prove identity continuity across prior transactions — the exact signal that flags a dispute as first-party rather than true fraud.

At Aurai, well-represented friendly-fraud cases win 75–85%. Without representment, they win 0%.

What actually reduces friendly fraud

Since prevention tools can't stop it, the reduction levers are behavioral:

1. Clear descriptors — half of "I don't recognize this charge" disputes are descriptor mismatches
2. Delivery notifications — post-purchase emails reduce non-receipt disputes
3. Frictionless refunds — customers who disputed instead of asking for a refund often did because refunding was hard
4. Subscription cancellation UX — trial-to-paid disputes turn into 13.2 disputes. Fix the UX
5. CE 3.0 systematically — structure identity-continuity data collection so representment is automatic