Bank Customer Churn Prediction

Dataset Overview

This dataset consists of 10,000 records with 12 columns related to bank customer information. The goal is to predict customer churn, where the target variable is the churn column (1 indicates the customer has churned, and 0 indicates they have not).

Columns:

1. customer_id: Unique identifier for each customer.
2. credit_score: The credit score of the customer.
3. country: Country of residence (e.g., France, Spain).
4. gender: Gender of the customer.
5. age: Age of the customer.
6. tenure: The number of years the customer has been with the bank.
7. balance: Account balance.
8. products_number: Number of products the customer uses.
9. credit_card: Whether the customer holds a credit card (1 = Yes, 0 = No).
10. active_member: Whether the customer is an active member (1 = Active, 0 = Inactive).
11. estimated_salary: Customer’s estimated salary.
12. churn: Whether the customer has churned (1 = Yes, 0 = No).

Project Description

In this project, I conducted an exploratory data analysis (EDA) and created visualizations to uncover patterns and insights related to customer churn. Key factors such as credit score, age, balance, number of products, and customer activity were analyzed to identify correlations with churn behavior. These insights are critical in helping banks strategize retention efforts and optimize their services.

Tools & Techniques:

• Data Cleaning: Managed missing or inconsistent data.
• Data Visualization: Utilized Power BI's built-in visualization capabilities to create insightful dashboards and reports.
• Feature Analysis: Focused on variables like age, balance, and tenure to understand customer profiles.

This analysis helps predict which customers are likely to churn, allowing banks to take preemptive actions to retain valuable clients.