Collection Analytics
Objective
To develop an easy to use statistical tool that scores active accounts in the portfolio on their propensities to default in the coming billing cycle.
Methodology
- Used historical profile, payment and last two years’ collections data to develop statistical models to predict future default behavior.
- Developed a MS Excel based tool with the model algorithms built into it.
- The tool is run on 1st day of every billing cycle; it scores all the accounts in the portfolio on their propensity to default.
- The scores are used for targeted collection treatments.
Impact
The collections team now targets only top 20% accounts with highest scores (90% of potential defaults are in top 20%). This minimizes collection expenses and delinquencies both.
Collection expenses have been reduced by 30% and monthly default rates have also gone down by 6% within six months of implementation.