Foundations · Fintech and Digital Finance

    Credit Scoring (Traditional and Alt-Data)

    8 min readLast reviewed: July 2025

    Intuition

    A credit score is a number that summarises an individual's credit history and predicts the probability of default on a future loan. Lenders use these scores to automate credit decisions, approve/reject, set interest rate, determine credit limit, at scale and speed impossible with manual underwriting.

    India's formal credit bureau system is relatively young compared to the US. The four SEBI-licensed credit information companies (CICs) are CIBIL TransUnion, Experian, Equifax, and CRIF Highmark. Each maintains a database of credit history for hundreds of millions of Indians, updated monthly by member banks and NBFCs.

    A major challenge in India is the thin-file problem: a large segment of the population (self-employed, rural, young first-time borrowers) have little or no credit history. Alternative data scoring, using mobile phone behaviour, utility payments, GST data, UPI transaction history, is being developed to extend credit access to this population.

    Mechanics

    CIBIL TransUnion Score:

    • Range: 300–900. Higher is better.
    • 750+ generally considered good for most lenders; 800+ for best rates
    • Factors (approximate weightage):
      • Payment history (35%): timeliness of past EMI/credit card payments
      • Credit utilisation (30%): how much of revolving credit limit is used
      • Credit history length (15%): how long accounts have been active
      • Credit mix (10%): diversity of secured (home loan, auto) and unsecured (personal, credit card) credit
      • New credit enquiries (10%): too many recent applications signal stress

    Bureau score vs internal scorecard:

    • Bureau score is based on your credit history with formal lenders
    • Most banks and NBFCs also build internal scorecards that combine bureau data with application data (income, employer, age, location) and sometimes behavioural data
    • Final credit decision typically uses both

    Alternative data scoring:

    • Mobile recharge frequency and pattern
    • Utility payment history
    • UPI transaction volume and regularity
    • GST filing and revenue data (SME)
    • Rental payment history
    • Educational credentials and employment data

    Try it yourself

    Interactive exercises coming soon.

    Key glossary terms

    Related topics

    Stay in the loop

    Roughly one email per month. No spam, no upsells.