NeoStats streamlined the targeting strategies of Digital leads by setting up likelihood scoring algorithms to take up a loan as well as assigning the right channel of communication and reducing the time gap in reaching out
Situation
The bank struggled with high drop-off rates and low conversion rates for digital leads on new personal loans. Inefficient tele-calling and time lags in customer outreach compounded the issue.
Challenges
- Very low digital penetration (98% drop offs)
- Inefficient use of resources & capacity (Reaching out to all drop offs via calling led to diluted CX, high operational cost, delay in contacting, tighter calling capacity and lower conversion)
- Limited response of existing contact strategies
Solution
Leveraging ML propensity models (Random Forest Classification), able to identify high-quality leads with higher chance of taking up new/top-up loans. Effective use of channel selection and targeted outreach significantly boosted conversions.
ETB
Propensity Model
NTB
Prioritization rules
Early drop offs
Identify ETBs & prioritization rules
