ThailandFinance and insurance2022 – ongoingSiam Commercial Bank
Machine learning for instant digital loan approval and product offers
Siam Commercial Bank doubled monthly digital loan bookings with machine learning behind instant approvals
Problem
The bank served more than 17 million customers and wanted to lend digitally without paperwork. It needed credit risk and eligibility predictions that worked at the moment of a customer's request.
Approach
The data team built predictive models for instant loan approval, risk assessment and collection strategy for high-risk customers. They also built models that predicted loan eligibility for existing customers without documents, and models that recommended investment and banking products. The team managed experiments and deployments with MLflow on Databricks.
Results
Takeaway
Pre-approving existing customers from data they already hold let the bank sell loans without a document step.
Categories
- Size
- Large enterprise
- Industry
- Finance and insurance
- Business function
- Finance, risk and legalMarketing and sales
- Innovation type
- Product and service innovation
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks reports that Siam Commercial Bank used predictive models for instant loan approval, eligibility prediction and product recommendations. It states that monthly new digital loan bookings doubled and that a new loan product brought in more than 300,000 customers in seven months.
Full details are in the original sources.
