A large services organization wanted to decrease customer attrition and reduce the associated customer reinstatement cost. The team of recovery specialists from this organization would call customers who have left the service and try to convince them to re-subscribe by providing incentives. The problem with this approach was that the recovery rate was very low and the customer reinstatement cost for the organization was high.
The organization wanted to identify customers who are at the highest risk of attrition and develop a proactive model for intervention to improve the retention rate at a reduced cost.
Data & Analysis
Our first step was to engage with the stakeholders and learn about their objectives and concerns, we then gathered insights related to customers that had terminated the service. What are the attributes of these customers? What was the reason for termination? What behavior and actions did customer take over a period of time? Which products and services they subscribed? How long they have been a customer and what is their profile? What competitor services they subscribed to before subscribing to our service? etc.
After careful data analysis of behaviors, patterns and profile data, the team developed a predictive analytics model that calculated likelihood of termination for every single customer. When a customer's likelihood of termination is above the threshold, the customer is considered as 'at-risk'.
The organization now has a customer dashboard for the 'at-risk' segment' and is tracking all the efforts made by the team as well as results achieved. This promising strategy and application of predictive analytics is already paying off with improved customer retention and lower reinstatement cost.