Credit scoring with alternative data
VioScore is an illustration of how credit scoring can use alternative data, meaning information about financial behaviour that does not come from a traditional credit file. It is written for readers in finance who want a clear, plain account of the idea.
Please note: everything on this site is illustrative. All data is invented and labelled SYNTHETIC. VioScore has not been validated, is not regulated or approved, and must not be used to make real lending decisions.
The problem
Traditional credit scores rely mainly on a person's record with lenders: loans, cards and repayments. People with little or no such record, sometimes called thin-file or credit-invisible, are hard to assess even when they pay their bills reliably. A lender may decline them or price their credit cautiously for lack of information, not for evidence of risk.
The idea of alternative data
Alternative data is information from outside the usual credit file that can still say something about how reliably a person meets regular obligations. Common examples include rent, utility and subscription payment history. Used carefully, with the person's consent, it can add a fuller picture for people who have little traditional history.
What VioScore shows
VioScore walks through a small, simplified scoring example using synthetic records. It shows how payment histories could be turned into a score and a band, and where human judgment still matters.
Care is needed
More data is not automatically better. Alternative data brings questions about fairness, privacy, consent and explainability, which the next page discusses. A score is an input to a decision made by people and policy, and not a substitute for them.
Where to go next
- How it works: what alternative data is, the steps from data to score, and the limits.
- Datasets: small synthetic tables and what each column means.