Traditional Credit Score Models

How traditional credit score models are created at companies like FICO, Equifax, etc.

Traditional credit scoring companies create empirically developed models that are tailored to

• product portfolios

• customer bases

• in-country credit bureau data

These custom models are built after extensive data analysis and generally provide a higher degree of performance.

Thus, they align the credit score to your risk environment, for example:

• Severity of negative behaviour (e.g., 60 days late versus charge-off)

• Duration of performance window (e.g., 12 versus 24 months)

• New or existing trade line performance (e.g., auto loans versus credit card)

• Regional specification: make adjustments to capitalize on differences in regions to better map lending footprint

Companies like FICO will not only build the data-driven score models to match your criteria using broad-based or unique data, but will also validate and maintain the models over time, ensuring that the score continues to perform in predicting and rank-ordering the credit risk of any consumer or SME, depending on the type of empirical score being developed.

“You’ll have your production model that you’re using today,” says Good from Credit Karma. “At the same time, [companies] are monitoring and back-testing to see how it performs against the vintage models that were previously used.”

And there are new models, sometimes called “challenger” models, that are tested against the current production, or “champion,” models.

Once a new model meets regulatory requirements and is validated, it can go into production alongside the current champion models.

“Sometimes it’s very subtle, and you’re just checking slight variations,” says Good. “You think you have some better data, maybe new bureau data or alternative data.”