📊 Customer Churn Analysis

First Data Science Project with Numpy,Pandas,Mitplotlib,Seaborn . Completely done by me , no AI help .

📌 Problem Statement

Businesses often lose customers without understanding the reason. This project analyzes customer data to identify patterns behind churn and suggest improvements.

---

📁 Dataset

• Source: Kaggle (Telco Customer Churn Dataset)
• Contains customer demographics, services, tenure, and billing details

---

🧠 Objectives

• Identify factors affecting customer churn
• Analyze customer behavior
• Provide actionable business insights

---

🛠️ Tools & Technologies

• Python
• Pandas
• Matplotlib
• Seaborn

---

🧹 Data Cleaning

• Handled missing values
• Converted data types
• Removed inconsistencies

---

📊 Exploratory Data Analysis (EDA)

Performed analysis on:

• Churn distribution
• Contract type vs churn
• Tenure vs churn
• Total charges vs churn
• Internet service vs churn

---

📈 Key Insights

• Customers with month-to-month contracts have the highest churn
• New customers (low tenure) are more likely to leave
• Customers with lower total charges tend to churn more
• Long-term contract users are more stable

---

🚀 Recommendations

• Encourage users to switch to long-term contracts
• Provide offers for new customers
• Improve onboarding experience
• Target at-risk users with retention strategies

---

📌 Conclusion

Customer churn is strongly influenced by contract type, tenure, and usage behavior. By focusing on early engagement and long-term commitment, businesses can reduce churn effectively.

---

📂 Project Structure

• Customer_Churn_Analysis.ipynb → Main notebook
• telco-churnData.csv → Real World Dataset (.csv)
• README.md → Project overview

---

✨ Future Improvements Targets

• Build a machine learning model to predict churn
• Deploy dashboard using Streamlit or Power BI

---

Author

MD. Abdur Rahim Ratul - Portfolio ( ratul.site )