Intelligence before the transaction

See your full fraud exposure.

A two-minute, model-calibrated estimate of detected fraud exposure — built from the one number every institution knows: your active digital user count. No fraud figures, no sensitive data required.

Two inputs are all you need to get started. You'll find the full method explained beneath your results.

Active digital users — last 12 months unique users

Use a consistent 12-month definition — the same basis you'd report internally.

Institution type

optional — refine with your own loss values

I know my average loss per fraud case

APP, USD

ATO, USD

3-D Secure, USD

Mule retail, USD

Mule corporate, USD

Leave blank to use the model's built-in values. Any figure you enter replaces only the loss-per-case component of the estimate.

Enter your active digital user count to run the estimate.

Your estimate appears here — two inputs is all it takes

Estimated customer fraud exposure · annual

—

APP + account takeover + online card fraud

Mule cash-out exposure — separate metric

—

Criminal funds moving through your accounts. Reported separately — never added to customer fraud.

Breakdown by fraud type · USD / year

What this number is: an estimate of fraud your institution would likely detect at this scale — not total fraud occurring, and not officially reported losses. It is a calibrated, order-of-magnitude figure for planning, not a confirmed loss.

How this is calculated ›

This estimate converts Group-IB's session-level fraud detection model into a per-user form, so it runs on one input every institution already knows.

1 · Detected fraud per session

How much detected fraud, by type, is associated with a typical digital session — derived from session-level modelling on live banking traffic.

2 · Sessions per user

A typical institution's average sessions per active user per year, by institution type.

3 · Your estimate

Combined into a per-user figure, then applied to your active user count.

The model is calibrated on real institutional data and represents a typical institution at your scale. Individual results vary with engagement and fraud patterns, so treat this as an order-of-magnitude estimate. Digital-wallet and digital-only estimates carry wider uncertainty than retail.

This is an estimate. See your real exposure.

A Group-IB Threat Exposure Assessment can build on this with a closer look at your own data — and at the underground-market exposure and unreported volume this estimate cannot see.

Request an assessment →