Wipro Ltd - Silent Attrition Prediction Model for Retail Banking
The Silent Attrition Prediction Model for Retail Banking was developed to identify the customers likely to churn silently within the next x months, to help retain valuable customers and to arrive at a foresight driven targeting strategy to generate better ROI.連絡合作夥伴 觀看影片
- Gold Application Development
- Gold Application Integration
- Gold Cloud Platform
- Gold Cloud Productivity
- Silver Data Platform
Silent Attrition happens when customers stop transacting without saying “I’m no longer your customer” or in other words, a customer stops patronizing a business without any explanation. Identifying silent attrition can be difficult because it requires analysis of the customer transaction history not just their current data status. However, countering silent attrition can be a very effective strategy to increase customer value because customers are already aware of your product or service. The approach this app takes to solve this problem is through a four-step assessment and analysis pattern starting with financial impact, transaction trends, predictive analysis and model evaluation. The data used includes customer details, behavioral data and POS/campaign data. The primary aspect is to analyze the financial impact caused by silent attrition customers and non-attrition customers based on dimensions like balance amount, card type, reward points and point-of-sale. Following this is the study of transaction trends with different views of Y-O-Y, monthly, quarterly and half yearly at an individual customer level for Top 10 customers. Having captured these two aspects, the predictive modelling thus helps, to identify the customers likely to silently churn w.r.t. risk category, card type, balance amount and the like. When combined with the response model, the high value, high response segments can be targeted to cut down marketing costs. It provides a foresight into possible churn from high value segments for proactive retention and helps target high value customers who are more responsive.