Reducing Customer Churn Leveraging Statistical Modeling and Predictive Analytics
The telecom giant was facing extreme customer attrition. It costs hundreds of dollars to acquire a new customer.
When a customer leaves, the company not only lose the future revenue from this customer but also the resources spent to acquire the customer in the first place. The senior management has decided to prevent the customer churn across all business divisions.
A top telecom organization based out of United States providing data, voice and managed services to large and mid-sized organizations.
They have the operations in the United States and are expanding the operations in Europe and Africa.
A team of Data Scientists, Data Engineers, and Business process consultants was formed and were tasked to identify the key parameters affecting customer churn for each business division.
Key Steps Taken To Identify Parameters of Churn
1. Exploratory analysis on satisfaction data - Installation, Fault, Communication etc. to determine the impact of different components of satisfaction data on the overall churn.
2. Built a predictive churn model leveraging SLA data (billing, delivery, and assurance), customer satisfaction data and complaints data to identify key drivers affecting churn
3. Separate models using techniques such as regression analysis were built for each business units and each service line, all rolling up to one overall churn model
Key Insights of Predictive Churn Model
1.A customer experiencing a delay in delivery was found to be 12 times more likely to churn, as compared to one who experienced an on-time delivery
2.Key parameters impacting churn for different divisions were very different. E.g. impact of value added services on churn was significant for large customers
3.The predictive churn model developed attained an accuracy range of 60-80%
The Data scientists team have now enabled advanced reports & dashboards to predict customers attiring from the business.By Unlocking insights, the management team initiated the right business decisions to prevent customer attrition.
Key Business Decisions Taken Based on Insights From Predictive Model
1.Develop sustainable and robust strategy for customer retention
2.Formulate plans to reacquire customers who have been moved to competitors
3.Convert low-revenue earning customers into highly profitable ones
4. Reduce customer defections and improve profits.
5. Track customer satisfaction by product, segment, and cost to serve.
Expected Savings due to Churn Prediction Model