Customer Churn in Financial Services

Columns

• Customer ID: A unique identifier for each customer.
• Surname: The customer's surname or last name.
• Credit Score: A numerical value representing the customer's credit score.
• Geography: The country where the customer resides, limited to France, Spain, or Germany.
• Gender: The customer's gender, either Male or Female.
• Age: The customer's age in years.
• Tenure: The number of years the customer has been with the bank.
• Balance: The customer's account balance.
• NumOfProducts: The number of bank products the customer currently uses (e.g., savings account, credit card).
• HasCrCard: Indicates whether the customer possesses a credit card (1 for yes, 0 for no).
• IsActiveMember: Indicates whether the customer is considered an active member (1 for yes, 0 for no).
• EstimatedSalary: The customer's estimated salary.
• Exited: The target variable, indicating whether the customer has churned (1 for yes, 0 for no).
• RowNumber: A sequential number for each row in the dataset.

Distribution

• RowNumber: Mean 5,000, Std. Deviation 2.89k, ranging from 1 to 10,000.
• Customer ID: Mean 15.7m, Std. Deviation 71.9k, ranging from 15.6m to 15.8m.
• Credit Score: Mean 651, Std. Deviation 96.7, ranging from 350 to 850.
• Age: Mean 38.9, Std. Deviation 10.5, ranging from 18 to 92.
• Tenure: Mean 5.01, Std. Deviation 2.89, ranging from 0 to 10 years.
• Balance: Mean 76.5k, Std. Deviation 62.4k, ranging from 0 to 251k. A notable proportion of customers have a balance of 0.
• NumOfProducts: Mean 1.53, Std. Deviation 0.58, ranging from 1 to 4.
• HasCrCard: Mean 0.71, Std. Deviation 0.46.
• IsActiveMember: Mean 0.51, Std. Deviation 0.5.
• EstimatedSalary: Mean 100k, Std. Deviation 57.5k, ranging from 11.6 to 200k.
• Exited: Mean 0.2, Std. Deviation 0.4.

Usage

• Developing machine learning models for customer churn prediction, particularly using techniques like Neural Networks and Random Forest.
• Data visualisation to explore patterns and relationships between customer attributes and churn behaviour.
• Data cleaning exercises as part of a machine learning workflow.
• Analysing customer behaviour in the banking sector to understand factors contributing to customer attrition.
• Building solutions for E-Commerce services related to customer retention strategies.

Coverage

• Geographic Scope: Customers primarily reside in France (most common at 50%), Spain, and Germany.
• Demographic Scope: Includes customer gender (55% Male, 45% Female) and age, with customers ranging from 18 to 92 years old.
• Time Range: The dataset does not specify a particular time range for the customer data.

License

Who Can Use It

• Data Scientists and Machine Learning Engineers looking to build and test churn prediction models.
• Business Analysts in the banking and financial services sector who aim to identify at-risk customers and inform retention strategies.
• Researchers studying customer behaviour, loyalty, and attrition in the financial industry.
• Students and educators for learning and teaching concepts in predictive analytics and customer relationship management.

Dataset Name Suggestions

• Bank Customer Churn Prediction Dataset
• Banking Churn Analytics
• Customer Churn in Financial Services
• Bank Client Retention Data

Attributes