Credit Card Users Churn Prediction.

Problem Statement

Thera Bank recently saw a steep decline in the number of users of its credit card. Credit cards are a good source of income for banks because of the different kinds of fees charged by the banks, such as annual fees, balance transfer fees, cash advance fees, late payment fees, foreign transaction fees, and others. Some fees are charged to every user irrespective of usage, while others are charged under specified circumstances.

Customers’ leaving credit card services would lead the bank to loss, so the bank wants to analyze the data of customers and identify the customers who will leave their credit card services and the reason for same – so that the bank could improve upon those areas

You as a Data Scientist at Thera Bank need to come up with a classification model that will help the bank improve its services so that customers do not renounce their credit cards.

Objective: To identify the underlying reasons for customer attrition and develop a predictive model to determine whether a customer was likely to discontinue using their credit card.

Data Description:

  1. CLIENTNUM: Client number. Unique identifier for the customer holding the account

  2. Attrition_Flag: Internal event (customer activity) variable - if the account is closed then "Attrited Customer" else "Existing Customer"

  3. Customer_Age: Age in Years

  4. Gender: The gender of the account holder

  5. Dependent_count: Number of dependents

  6. Education_Level: Educational Qualification of the account holder - Graduate, High School, Unknown, Uneducated, College(refers to college students), Post-Graduate, Doctorate

  7. Marital_Status: Marital Status of the account holder

  8. Income_Category: Annual Income Category of the account holder

  9. Card_Category: Type of Card

  10. Months_on_book: Period of relationship with the bank (in months)

  11. Total_Relationship_Count: Total no. of products held by the customer

  12. Months_Inactive_12_mon: No. of months inactive in the last 12 months

  13. Contacts_Count_12_mon: No. of Contacts in the last 12 months

  14. Credit_Limit: Credit Limit on the Credit Card

  15. Total_Revolving_Bal: Total Revolving Balance on the Credit Card

  16. Avg_Open_To_Buy: Open to Buy Credit Line (Average of last 12 months)

  17. Total_Amt_Chng_Q4_Q1: Change in Transaction Amount (Q4 over Q1)

  18. Total_Trans_Amt: Total Transaction Amount (Last 12 months)

  19. Total_Trans_Ct: Total Transaction Count (Last 12 months)

  20. Total_Ct_Chng_Q4_Q1: Change in Transaction Count (Q4 over Q1)

  21. Avg_Utilization_Ratio: Average Card Utilization Ratio

Data Cleaning and Feature Engineering Process:

  1. Handling Missing Values:

    • Numerical Variables: Missing values were replaced with the median of the respective columns using SimpleImputer(strategy="median").

    • Categorical Variables: Missing values were replaced with the most frequent value in the respective columns using SimpleImputer(strategy="most_frequent").

  2. Feature Transformation:

    • Numerical Features: A list of numerical features was created, including columns like Customer_Age, Credit_Limit, and Total_Trans_Amt.

    • Categorical Features: Categorical features such as Gender, Education_Level, and Income_Category were identified for further processing.

    • Categorical variables were one-hot encoded to convert them into a format suitable for machine learning algorithms using OneHotEncoder(handle_unknown="ignore").

Business Recommendations:

We developed a predictive model that the bank can deploy to identify customers at risk of attrition and understand the key factors driving customer churn. This model can guide the bank in implementing effective retention strategies and policies. Key factors influencing attrition include the number of transactions, revolving balance, transaction amount, and the total relationship count. Customers with fewer transactions (Total_Trans_Ct) are more likely to leave, so the bank can introduce incentives like cashback or special discounts to encourage card usage. Similarly, customers with lower revolving balances may have cleared their dues and chosen to opt out of credit card services. By seeking feedback from these customers, the bank can address the underlying reasons for their departure. Additionally, customers with lower transaction amounts (Total_Trans_Amt) may benefit from targeted promotions on higher-value purchases, which could increase transaction volumes and improve satisfaction.

The analysis also reveals other actionable insights. Customers using only one or two products (Total_Relationship_Count) account for the majority of attrition, suggesting a need to investigate potential issues with these products or improve customer support and transparency. Female customers, who generally utilize more credit and make higher-value transactions, should be prioritized for marketing campaigns. Offering increased credit limits to these customers could benefit both the bank and its clientele. Inactivity for 2-4 months is a major contributor to attrition, so the bank should proactively engage customers through automated reminders about offers, services, or activity summaries. Furthermore, customers with frequent interactions with the bank are also leaving, indicating dissatisfaction with problem resolution. To address this, the bank should implement a robust feedback collection system to ensure customer issues are resolved effectively and improve satisfaction to reduce attrition.