Credit Card Customer Attrition Prediction Data

Columns

• CLIENTNUM: A unique identifier for each customer. (Integer)
• Attrition_Flag: A flag indicating whether the customer has churned out. (Boolean)
• Customer_Age: The age of the customer. (Integer)
• Gender: The gender of the customer. (String)
• Dependent_count: The number of dependents the customer has. (Integer)
• Education_Level: The education level of the customer. (String)
• Marital_Status: The marital status of the customer. (String)
• Income_Category: The income category of the customer. (String)
• Card_Category: The type of card held by the customer. (String)
• Months_on_book: How long the customer has been on the books (i.e., with the provider). (Integer)
• Total_Relationship_Count: The total number of relationships the customer has with the credit card provider. (Integer)
• Months_Inactive_12_mon: The number of months the customer has been inactive in the last twelve months. (Integer)
• Contacts_Count_12_mon: The number of contacts the customer has had in the last twelve months. (Integer)
• Credit_Limit: The credit limit of the customer. (Integer)
• Total_Revolving_Bal: The total revolving balance of the customer. (Integer)
• Avg_Open_To_Buy: The average open to buy rate of the customer. (Integer)
• Total_Amt_Chng_Q4_Q1: The total amount changed from quarter 4 to quarter 1. (Integer)
• Total_Trans_Amt: The total transaction amount. (Integer)
• Total_Trans_Ct: The total transaction count. (Integer)
• Total_Ct_Chng_Q4_Q1: The total count changed from quarter 4 to quarter 1. (Integer)
• Avg_Utilization_Ratio: The average utilisation ratio of the customer. (Integer)
• Naive_Bayes_Classifier_Attrition_Flag_Card_Category_Contacts_Count_12_mon_Dependent_count_Education_Level_Months_Inactive_12_mon_1: A Naive Bayes classifier for predicting churn based on specific characteristics. (Integer)
• Naive_Bayes_Classifier_Attrition_Flag_Card_Category_Contacts_Count_12_mon_Dependent_count_Education_Level_Months_Inactive_12_mon_2: A second Naive Bayes Classifier Attrition Flag based on similar characteristics. (Integer)

Distribution

Usage

• Analyse key factors that influence customer attrition.
• Understand customer demographics, spending patterns, and their relationship with the credit card provider to better predict customer attrition.
• Determine which customer demographic is more likely to churn, using variables such as gender, marital status, education level, and income category.
• Analyse customer spending behaviour leading up to churning to predict the likelihood of a customer churning in the future.
• Create a classifier that can predict potential customers more susceptible to attrition based on their credit score, credit limit, utilisation ratio, and other spending behaviour metrics over time, serving as an early warning system for predicting attrition.

Coverage

License

Who Can Use It

• Data analysts looking to predict customer attrition in credit card portfolios.
• Researchers interested in customer behaviour, churn analysis, and predictive modelling in financial services.
• Financial institutions and banks aiming to manage their customer portfolios effectively and implement retention strategies.
• Data scientists developing machine learning models for early churn detection and customer segmentation.

Dataset Name Suggestions

• Credit Card Customer Attrition Prediction Data
• Bank Customer Churn Analytics Dataset
• Credit Card Holder Behaviour & Attrition
• Customer Loyalty in Credit Card Portfolios

Attributes