New Account Fraud: Stop Fake Customers Before They Create Chargebacks

New account fraud happens when someone opens an account using a stolen or fabricated identity, then uses that account to place orders that later get disputed, refunded, or written off once the real cardholder or a synthetic identity's true owner surfaces. It matters to merchants because the loss rarely stops at the fraudulent order: it usually resurfaces weeks later as a chargeback you have almost no evidence to fight. Machine-learning screening, the same techniques behind how AI detects fake accounts on ecommerce platforms, catches a meaningful share of these signups before the first fraudulent order ships.

The FTC's IdentityTheft.gov received more than 1.1 million identity theft reports in 2024, and credit card fraud, including new accounts opened in someone else's name, was the single most commonly reported category. Separately, TransUnion's H1 2026 fraud trends update found that 8.3% of digital account creation attempts globally were suspected of fraud in 2025, making account signup the highest-risk moment in the entire customer lifecycle. Neither number is shrinking, which is why the fix has to start before the fraudulent account ever places an order, not after a chargeback arrives, and it has to sit inside a broader ecommerce fraud prevention program rather than stand alone as a signup gate.

Following the Attack From Signup to Dispute

New account fraud rarely stops at account creation. It moves through a predictable sequence, and each stage leaves signals worth capturing:

• Account creation: a fraudster registers using stolen personal information (identity theft) or a blend of real and fabricated details (synthetic identity fraud).
• First transaction: the new account places an order, often testing with a small purchase before a larger one, or going straight for high-value, easily resold goods.
• Fulfillment: the order ships before any red flag has time to surface, especially if velocity and address-mismatch checks were skipped at signup.
• Dispute: the real cardholder (in identity theft) or the issuer (once a synthetic identity's credit profile collapses) files a dispute, and the merchant is left defending an order placed by an account it never verified.

Two patterns get lumped together under "new account fraud" but behave differently, and treating them the same weakens your response:

Signals, Controls, and the False-Positive Tradeoff

Every control that catches new account fraud also has a false-positive cost, and the two have to be weighed together rather than treated as a pure win:

• Velocity checks: multiple accounts from the same IP, device, or payment method in a short window. High signal, but shared households, corporate networks, or app-based traffic can trigger false flags.
• Identity and address consistency: mismatched billing and shipping addresses, or geolocation that doesn't match either. Strong for catching stolen-identity orders, weaker against synthetic identities that were built with internally consistent fake data.
• Device fingerprinting: reused devices across accounts that shouldn't be related. Effective against repeat fraud rings, less useful against one-off identity theft.
• Step-up authentication: requiring stronger verification at signup or first purchase. Reduces fraud but adds friction that can cost legitimate signups if applied indiscriminately.

None of these controls change dispute liability on their own. A transaction that passed every check can still be disputed by the real identity owner, which is why the evidence you capture at signup and checkout matters as much as the controls that screen for risk in the first place.

What to Retain Before and After Fulfillment

When a new-account order turns into a dispute, the merchant's position depends entirely on what was captured at the time of the transaction, not what can be reconstructed afterward. Retain:

• Account creation timestamp, IP address, and device fingerprint.
• Any identity or address verification performed at signup, and its result, including how your payment service provider scored the transaction.
• Billing-to-shipping address match status at checkout.
• Authentication result if 3DS or another card-not-present fraud control was used on the transaction.
• Delivery confirmation, including signature or photo proof where available, tied to the account and order.

This is the same underlying discipline behind compelling evidence for any dispute type: capture it automatically at the moment of the transaction, because reconstructing it after a chargeback notice arrives is far harder and often impossible.

Closing the Loop With Dispute Results and Reason Codes

The dispute outcome itself is a signal worth feeding back into your fraud controls, not just a loss to absorb. If new-account orders are consistently disputing under fraud-related chargeback fraud reason codes, that pattern points to a gap in signup verification, not checkout friction. If they're disputing as item-not-received or not-as-described, the real issue may be first-party friendly fraud riding in on a new account rather than a stolen identity, which calls for a different evidence response entirely.

Treating chargeback fraud prevention and new account screening as one continuous system, rather than a fraud team and a disputes team working from separate data, is what actually shortens the gap between a bad signup and a stopped one. Chargeback prevention alerts add another layer by flagging a dispute before it fully posts, giving you a window to intervene on a new-account order before it becomes a loss on the books.

Prevention Only Pays Off When It's Measured in Recovered Disputes

A fraud control that blocks new account fraud but can't be tied back to fewer chargebacks is hard to justify against the false-positive cost it adds. The stronger approach connects pre-transaction screening directly to what happens when a dispute lands: the same signals that flagged an account as risky at signup are exactly what should show up in the evidence response if that account still slips through and disputes later.

Frequently Asked Questions

What is new account fraud?

New account fraud is the use of a stolen or fabricated identity to open an account, then place orders that are later disputed, refunded, or written off once the real identity owner or the issuer identifies the fraud.

What is the difference between new account fraud and account takeover?

New account fraud creates a new account using a stolen or synthetic identity. Account takeover instead hijacks an existing, legitimate account. Both can lead to fraudulent orders, but the controls that catch them differ because one involves an unfamiliar identity and the other involves a compromised familiar one.

How common is new account fraud?

TransUnion's H1 2026 fraud trends update found that 8.3% of digital account creation attempts globally were suspected of fraud in 2025. The FTC separately received more than 1.1 million identity theft reports in 2024, with credit card fraud, including fraudulent new accounts, as the most reported category.

Does verifying identity at signup stop chargebacks?

It reduces the volume of fraudulent accounts that reach checkout, but it doesn't eliminate dispute risk. A transaction that passes every signup check can still be disputed later, which is why retaining verification and transaction evidence matters alongside prevention controls.

What is synthetic identity fraud?

Synthetic identity fraud combines real information, such as a stolen Social Security number, with fabricated details to build an identity that doesn't belong to any real person. It's harder to detect than stolen-identity fraud because the resulting profile can look internally consistent.

Chargeflow connects your pre-transaction fraud signals to automated evidence and dispute recovery through Chargeflow Prevent, so a new account that slips past screening doesn't also slip past your defense.