Telecom Customer Churn Analysis
Python - Power BI - Customer Retention
A customer churn analysis project identifying the service, support, contract, and customer-profile factors most associated with telecom attrition.
Problem
Knowing the churn rate tells you nothing you can act on. The business question is which of service, contract and support behaviour actually moves it, and which of those you can change.
Approach
Python-driven EDA across the customer base to isolate the service, contract type, tenure and support factors most associated with leaving, then KPI design to turn those into measures worth tracking rather than one-off findings.
Trade-off
The output is a Power BI dashboard and a written recommendation rather than a predictive model. A churn classifier would have been the more impressive artefact, but the levers here are contract and support decisions, and that needs something the retention team can read.
Architecture
Analysed a telecom dataset with Python-driven EDA and KPI design, then translated the findings into a Power BI dashboard and written recommendations for retention strategy, service improvement, and churn reduction.
Results
Retention levers surfaced across service, contract and support behaviour, delivered as a dashboard plus written recommendations aimed at the people who would act on them.
Highlights
- Customer churn drivers
- Retention dashboard
- Recommendation report
Technologies
- Python
- Power BI
- EDA
- KPI Design
- Customer Analytics
- Retention Analysis
Screenshots
