Analytics and AI
Customised analytical models implementation for a large bank
- Ineffective campaign/cross-sell & up-sell revenues due to:
- Absence of appropriate segmentation.
- Lack of insights into customer spending patterns.
- Inability to predict bank trade utilization potential.
- Limited insights into CASA customers.
- Lack of campaign efficacy analysis.
- Unknown affluent category customers.
- AML: Siloes in the process of AML mule account detection leading to sub-optimal regulatory reporting to the central bank.
Tailored analytical models implementing supervised and unsupervised learning algorithms solving specific business problems around customer insights, cross-sell/up-sell, campaign inefficiency and regulatory reporting.
- Customer segmentation with high and low impact aided in targeting a specific group of customers for campaigns in terms of shifts between the Customer segments e.g., shift from “At Risk” to “Potential Loyalist” or from “Potential Loyalist” to “Champions” and the vice versa.
- Customer Journey Analysis aided in predicting the customer spend behavior every 3 months and when the customer is likely to change spending pattern.
- Improved bank-trade utilization by predicting potential customer base for bank-trade.
- Higher insights into monthly incomes of CASA customers based on their spending and their investment.
- Campaign efficacy analysis through pre and post campaign ROI, segmentation-based most recent spending.
- Re-classification of affluent customers for inclusion of customers who are affluent but not classified as affluent.
- Improved regulatory (AML) reporting to the central bank through automated process of detection of suspected mule accounts from savings accounts data and generation of main mule leads for verification by AML team.