Behavioural analytics
Behavioural analytics in fraud detection compares a transaction with what the account has done before, using a baseline built from its own history instead of a limit applied to everyone. A transaction that departs sharply from the baseline is riskier even when it breaches no fixed threshold.
In more detail
The value of a baseline is that it is specific. Five thousand euro is unremarkable for a business account and highly unusual for a salary account that has never sent more than four hundred. One fixed threshold cannot express that difference, and a threshold low enough to protect the salary account would drown the business account in alerts.
The cost is that a baseline needs history. A newly opened account has none, so behavioural checks are normally paired with rules aimed at new accounts, and the baseline has to be allowed to drift as a customer's life changes.
The system calculates the baseline from past events, then scores each new transaction by how far it sits from it.
Example
An account that usually pays small amounts to a handful of regular payees at weekday lunchtimes sends a large amount to an unfamiliar payee at three in the morning. No fixed limit is breached, but every feature differs from the account's norm.
Related capability
Behavioural monitoring on the platform
Related terms
Discuss your transaction monitoring requirements
Tell us which transaction flows you need to monitor, which systems Akoben would integrate with, and how your fraud team works today.
Request a demo