A case study on churn analysis1
Download as PPTX, PDF2 likes2,282 views
The document discusses a case study report on churn analysis for a telecom service provider. It outlines the business scenario of a telecom provider losing customers and profitability with average churn rates of 8%, 12%, and 15% over three quarters. The solution proposed uses advanced modeling techniques like neural networks and logistic regression to construct a model that scores each customer's probability of churn. The model helps identify high-value customers likely to churn and informs the provider's retention strategy.
A case study on churn analysis1
- 1. Axion Connenct 1 A CASE STUDY REPORT ON CHURN ANALYSIS Submitted to Mr. Sanjay Rao Founder, Axion connect Presented by Amit Kumar
- 2. Way Forward…. 2 A Business Scenario Business Problem Available Solution Stepwise solution Results & Findings
- 3. A Business Scenario… 3 A leading telecom service provider has a customer base of 1million users. In the cellular base, a customer can choose pre-paid & post-paid services. In this competitive telecom market customers have vast array of choices, the cost of acquisition & rate of customer churn, both are increasing at a rapid pace. In last three quarters operator’s profitability has gone down & faced problem of customer churn in one of the operator’s largest circle. The average attrition rate for each quarters for the operator is 8% , 12% & 15% respectively. Axion Connenct
- 4. Business Problem.. 4 Telecom service provider is loosing customer base & their profitability has gone down. The average churn rate is around 12%(Q1-8%, Q2-12%, Q3-15%) The acceptable return on their retention program is very less & it has not targeted sharply to the customer churned in second & third quarters Axion Connenct
- 5. Available solution.. 5 To address these problems, operator wants a robust retention model/churn model that would help the telecom operator to identify the propensity of churn & high-value customers. Need to use advance modeling techniques like neural networks, decision trees & logistic regression to construct a model that can score each customer for his probability of churn over next quarter. Axion Connenct
- 6. Stepwise solution.. 6 Integration of the data like, billing information, demographic information, service record information, customer participation in retention program etc. in a single file to capture all aspects of customer interaction. Understanding of data dimension, functions & association of data object with functions Data object preparation Constructing the model Validating the model Implementing it & track it Axion Connenct
- 7. Results & Findings.. 7 Model will help the operator with other CRM metrics to build retention strategy. Customer with high profitability with high propensity to churn should be in the highest priority of retention and should offer best incentive. Customer with low profitability with high propensity to churn should encouraged to increase the usage. Axion Connenct
- 8. Axion Connenct 8 THANK YOU !