Customer Winback Using Product Hook Strategy
For a Large Asset Management Company in India.
CLIENT & PROBLEM STATEMENT
- The client is a major asset management company in India with diversified product mix and a large customer base.
- The client wanted to reactivate inactive clients who had not conducted any transactions in the past five years. The objective was to devise a strategy to win back these dormant clients and encourage them to resume their engagement with the company.
APPROACH
- To address the client’s goal we focused on product-wise analysis to identify the most effective schemes for re-engagement. Utilizing XGBoost classifier, we developed predictive ML models capable of identifying the best buying products for each target customer.
- We improve the model quality using accuracy measures such as Gini, sensitivity, specificity, and others. This ensured that the models were robust and capable of accurately predicting customer behavior.
- Our approach effectively identifies and target dormant clients for re-engagement. By leveraging ML modeling, we identified the most suitable schemes for upselling and cross-selling to dormant clients, maximizing the likeliho
SOLUTION & OUTPUT
- Using predetermined cutoffs based on sensitivity, specificity, and target customers, we tagged dormant clients and recommended specific schemes tailored to their preferences and investment history.
- Once the models were trained and validated, we proceeded to identify the list of existing customers who could be targeted for winback.
- As a result of this initiative, revenue increased by 7%.