Behavioral score
A credit score built from how an existing customer uses the account, refreshed regularly, unlike a score set once at application.
Category
How a model is chosen and adapted
Related terms
A behavioral score estimates an existing customer's credit risk from how the account is actually used: utilization and its trend, payments, days past due, inflows, and bureau changes, refreshed on a monthly or weekly schedule.
How it differs from an application score
An application score is set once, from what is known when the customer applies. A behavioral score keeps moving with the customer, so it can show that a good customer has outgrown a limit or that a performing one is quietly deteriorating. It is built and validated like any other credit model, as a scorecard or a constrained machine-learning model, with reason codes for each band.
What it feeds
Limit reviews, drawdown answers on revolving lines, and early warning. The score is an input; the credit policy says what each band allows, and changes outside the pre-authorized cells go to an underwriter. The behavioral score and limit management use case shows the whole path.
Want these terms worked through on one of your own decisions? Start with a Decision Blueprint