Customer Churn Analysis

This project focuses on performing Exploratory Data Analysis (EDA) on telecom customer data to understand customer churn behavior. The dataset is sourced from Kaggle and contains customer demographics, account information, and service usage details.

🔗 Dataset Source: Kaggle - Telco Customer Churn

---

Dataset Overview

• Rows: 7,043
• Columns: 21
• Target Variable: Churn (Yes/No)
• Contains both categorical and numerical features.

---

Project Steps

1. Importing and Loading Data

• Loaded dataset using pandas.

2. Data Cleaning

• Dropped irrelevant customerID column.
• Converted TotalCharges to numeric type.
• Handled missing values by dropping rows with nulls.

3. Feature Engineering

• Mapped Churn values:
  • Yes → Churned
  • No → Retained

4. Data Exploration

• Checked data types and basic statistics.
• Created separate DataFrames for Churned and Retained customers.
• Analyzed churn distribution, tenure, and billing patterns.

---

Key Insights

• Customers with shorter tenure are more likely to churn.
• Month-to-month contracts show a higher churn rate.
• Higher monthly charges may be correlated with churn.
• Long-term contract types (1- or 2-year) have lower churn rates.

---

Technologies Used

• Python
• Pandas
• Jupyter Notebook