Telecom Customer Churn Analysis & Dashboard 🚀 Project Overview
This project is an end-to-end analysis of telecom customer churn, combining Python (Pandas & ML) for data analysis and Power BI for dashboard visualization.
The goal is to understand customer behavior, predict churn, and provide actionable business insights for customer retention. ############################################# Dataset
Source: Kaggle (Telco Customer Churn Dataset) #################################### Features include:
Customer demographics: Gender, SeniorCitizen, Partner, Dependents
Service information: InternetService, Contract, PaymentMethod, PhoneService
Charges: MonthlyCharges, TotalCharges
Target: Churn (Yes/No → 1/0)
Dataset cleaned and missing values handled in Python. ############################### Tools & Technologies
Python Libraries: Pandas, NumPy, Matplotlib, Seaborn, scikit-learn
Machine Learning: Logistic Regression (baseline churn prediction)
Power BI: Interactive dashboards with KPIs, charts, and slicers
#######################################################
Project Steps
Data Cleaning & Preprocessing
Handled missing values and datatype conversions
Encoded categorical variables (Yes/No → 1/0)
Exported cleaned dataset for Power BI
Exploratory Data Analysis (EDA)
Churn distribution (bar chart)
Monthly Charges vs Churn (Pie chart showing revenue)
Customer Tenure distribution (binned column chart)
Senior citizen churn analysis
Churn by Internet Service (donut chart)
#######################################################
Machine Learning Model
Logistic Regression to predict customer churn
Evaluation metrics: Accuracy (81%), Confusion Matrix, Classification Report
Insights: Higher churn for month-to-month contracts, fiber optic customers, high monthly charges, and new customers
#############################################
Power BI Dashboard
Interactive visuals with slicers: Gender, SeniorCitizen, InternetService, Contract, PaymentMethod
KPIs: Total Customers, Churn Rate, Total Revenue
Charts: Churn by Contract, Revenue Pie Chart, Tenure Distribution, Internet Service Churn
#################################################
Key Insights
Month-to-month customers churn the most
Higher monthly charges → higher churn
Fiber optic customers churn more than DSL
Most churn happens within the first 12 months
Senior citizens show slightly higher churn
#########################################
Project Structure Telecom_Churn_Project/ │ ├── Telecom_Churn_Analysis.ipynb # Jupyter Notebook with full Python code ├── cleaned_churn.csv # Cleaned dataset ├── PowerBI_Dashboard_Screenshots/ # Optional folder with dashboard images └── README.md # This file
#######################################
Skills Demonstrated
Data Cleaning & Preprocessing
Exploratory Data Analysis (EDA)
Logistic Regression for churn prediction
Data Visualization & Dashboard Creation in Power BI
Business Insights & Storytelling