Behavioural analytics
Also called behavior analytics, behavioural monitoring, customer behaviour profiling
Behavioural analytics is the use of data on how a customer normally acts to spot activity that does not fit. It builds a baseline from payment amounts, timing, payees, devices and login habits, then scores each new event against it. Large deviations generate alerts for fraud or money laundering review.
How it works
The system builds a profile for each customer and for peer groups of similar customers. A payment to a new payee at an unusual hour from a new device scores higher than a routine bill. Fraud teams use the score in real time to hold or challenge a payment. AML teams use it to flag accounts whose activity drifts from their stated purpose.
Behavioural signals catch scams and mule activity that fixed rules miss, including authorised push payment fraud where the customer is being manipulated. EU and UK payment rules expect firms to monitor transactions for abnormal behaviour, and low fraud rates let firms skip strong authentication on some payments. Models need validation and monitoring for drift as customer habits change.
The rules
Requires transaction monitoring mechanisms that detect unauthorised or fraudulent payments, and allows risk-based exemptions where real-time analysis shows low risk.
Payment service providers in the EURequires sending and receiving firms to share reimbursement of authorised push payment scam losses, raising the value of behavioural scam detection.
Payment firms sending Faster PaymentsCommon questions
How is behavioural analytics different from biometrics?
Biometrics confirm who a person is, using face, voice or fingerprint. Behavioural analytics looks at how an account is used. Behavioural biometrics, such as typing rhythm, sit between the two.
Can behavioural analytics stop authorised push payment scams?
It can flag payments that break a customer's normal pattern, such as a large first payment to a new payee. Firms then add friction, such as warnings or a call, before releasing funds.
Does behavioural analytics reduce false positives?
It can, because alerts reflect each customer's own normal activity rather than one threshold for all. Results depend on data quality and on how well models are tuned and validated.
What data does it use?
Typically transaction amounts and timing, counterparties, channels, device and location data, and login patterns. Firms process this personal data within the limits set by data protection law.
Our coverage
News and analysis on transaction monitoring from RegTech Post.
Read the coverage →RegTech PostBiometric KYC Comes of Age: The Next Frontier in Digital IdentityRead the article →RegTech PostForter Quietly Rewrites the Rules of Online TrustRead the article →
