Project Type: Customer Retention & Business Intelligence
Telecom companies face high customer churn rates, leading to significant revenue loss. This project analyzes a dataset of 7,043 customers to identify the primary drivers of churn—such as contract types, high monthly charges, and technical support gaps. By leveraging Power BI and Excel, this analysis delivers actionable insights and interactive dashboards to help stakeholders identify at-risk customers and implement data-driven retention strategies.
File: Customer-Churn-Dataset.xlsx The dataset contains 7,043 records across 23 attributes, providing a 360-degree view of the customer profile:
| Category | Attributes |
|---|---|
| Demographics | customerID, gender, SeniorCitizen, Partner, Dependents |
| Tenure & Services | tenure, PhoneService, MultipleLines, InternetService, OnlineSecurity, OnlineBackup, DeviceProtection, TechSupport, StreamingTV, StreamingMovies |
| Billing & Contract | Contract, PaperlessBilling, PaymentMethod, MonthlyCharges, TotalCharges |
| Support Activity | numAdminTickets, numTechTickets |
| Target | Churn (Yes/No) |
Data Cleaning:
- Handled missing values in
TotalCharges. - Created tenure buckets (e.g., 0–1 year, 1–2 years) for better segmentation.
- Validated Churn KPIs: 1,869 Churned vs. 5,174 Retained.
- Data Preparation: Used Excel to profile the data, handle missing values, and calculate baseline KPIs (Total Customers: 7,043, Churn Rate: 26.5%).
- Interactive Dashboards: Developed two primary views in Customer-Churn-Dashboard.pbix:
- Churn Overview: High-level KPIs focusing on demographic splits and service usage.
- Customer Risk Analysis: A deep dive into behavioral triggers like payment methods, contract length, and support tickets.
- DAX Implementation: Created dynamic measures for Churn Rate %, Average Monthly Charges, and Ticket Frequency to allow for real-time filtering via slicers.
- Overall Churn Rate: 26.5% (1,869 out of 7,043 customers).
- Financial Impact: Churned customers have a higher Average Monthly Charge ($74.44) compared to retained customers ($61.27), indicating that price sensitivity is a major factor.
- Contract Type: Customers on Month-to-Month contracts are significantly more likely to churn (~43%) compared to those on one or two-year plans.
- Payment Method: Electronic Check users show the highest churn rate (~45%), while customers on Automatic Credit Card or Bank Transfers are the most loyal (~16–17%).
- Fiber Optic Paradox: Despite being a high-speed service, Fiber Optic users churn at a rate of ~42%, nearly double the rate of DSL users.
- Technical Friction: Churned customers averaged a higher number of Tech Support Tickets. Unresolved technical issues are a primary indicator of "imminent churn."
- Senior Citizens churn at a higher rate (approx. 41%) compared to younger demographics.
- Gender does not play a statistically significant role in churn decisions.
- Payment Migration: Offer a small one-time discount (e.g., $5 credit) for customers who switch from Electronic Check to Auto-pay methods.
- Fiber Optic Retention: Investigate the Fiber Optic service quality or pricing structure, as this high-value segment is currently the most unstable.
- Proactive Support: Use the "Tech Ticket" metric as an early warning system. If a customer opens >2 tech tickets in a month, trigger an automated "Customer Success" follow-up.
- Contract Incentives: Target month-to-month users with "Loyalty Upgrades" to move them into 1-year contracts, reducing the churn risk significantly.
| Tool | Purpose |
|---|---|
| Power BI | Dashboards, DAX measures, slicers, and interactive data visualization |
| Excel | Initial dataset loading, data cleaning, profiling, and calculation of baseline KPIs |
This project demonstrates the ability to transform raw telecom data into a strategic roadmap for customer retention.