Behavioural Scoring Runs After The Account Opens

Application scoring answers whether to lend. After that, a second kind of model takes over, built on what the customer actually does, and it governs most of what happens next.

Direct observation beats inference

An application score infers likely behaviour from a population. A behavioural score observes this customer: payment timing, amounts, balance trends, transaction types and how the account responds to changes.

Direct observation is more predictive, so once a few months of history exist, the internal score generally carries more weight than the original assessment.

This is why a lender's treatment of an existing customer can diverge sharply from what the credit file alone would suggest.

The score drives ordinary decisions

Limit increases, promotional offers, renewal terms and pricing at repricing points are typically driven by the behavioural score rather than by a fresh application assessment.

Because these decisions are made internally, they usually generate no search footprint, and the customer sees an outcome without an obvious trigger.

The same score decides which accounts are left alone and which are reviewed when the lender wants to reduce exposure.

Collections treatment is segmented by it

When an account falls behind, the response is not uniform. Behavioural data separates customers likely to self-cure from those likely to deteriorate further.

Accounts in the first group may receive a reminder and little else, while those in the second are contacted sooner and offered arrangements earlier.

The segmentation is about expected outcomes rather than about the amount owed, which is why two similar arrears positions are handled differently.

Patterns matter more than single events

A single late payment on an otherwise consistent account moves an internal score less than a gradual change in pattern: balances creeping up, payments dropping towards the minimum, cash usage appearing.

Trends are read as leading indicators, since they usually precede arrears rather than following them.

Customers who have never missed a payment can therefore see limits reduced or offers withdrawn, which reads as arbitrary from outside.

Bureau data still feeds the picture

Lenders commonly supplement internal data with periodic reviews of credit file information, to see commitments held elsewhere and arrears the account itself would not reveal.

Those reviews are conducted under the permissions in the agreement, and what they may be used for differs between jurisdictions.

The combination gives the lender a fuller view of its customer than any single credit file provides, and it is refreshed continuously rather than at application.