This project focuses on analyzing customer churn behavior in a telecommunications dataset.
Customer churn refers to the percentage of customers who stop using a company's service during a given period.
- Dataset Size: 7,043 customers
- Features: 21 columns
- Includes: Demographics, subscription details, billing info, churn status
🎯 Goal:
Identify patterns and factors that lead to customer churn and provide insights to improve customer retention.
Steps performed during analysis:
- Checked dataset shape and column details
- Handled missing and inconsistent values
- Converted categorical data into usable format
- Prepared dataset for analysis and visualization
✔️ Ensured clean and structured data for better insights
- ✅ 73% customers stayed
- ❌ 27% customers churned
📊 Insight:
A significant portion of customers are leaving, which can impact business growth.
- Most customers are non-senior citizens
- However, senior citizens have higher churn rate
📊 Insight:
Different customer groups behave differently and need targeted strategies.
- 🔴 Month-to-Month → Highest churn
- 🟡 One-Year → Moderate churn
- 🟢 Two-Year → Lowest churn
📊 Insight:
Long-term contracts improve customer retention.
- Pie Charts → Churn distribution
- Count Plots → Category comparison
- GroupBy Analysis (Pandas)
- Percentage-based comparisons
✔️ Helped in understanding patterns clearly
- Around 27% customers churned
- Month-to-Month users are high-risk customers
- Long-term contracts increase retention
- Customer demographics affect churn
- Data visualization helps identify patterns easily
This project shows how data analysis and visualization can help understand customer behavior.
- Encourage long-term contracts
- Target high-risk customers
- Improve service quality using data insights
- Python 🐍
- Pandas
- Matplotlib
- Seaborn
👩💻 Priya Pandey
B.Tech (3rd Year)
Give it a ⭐ on GitHub!