Feature-Oriented Campaign Targeting Model for Credit Card Customer Segmentation

Abstract

Rising competition in retail banking requires data-driven, interpretable targeting for credit-card campaigns by considering big customer related to considered customer features. On the other hand, the data gathering process is another crucial problem to handle since the system’s complexity is exponentially enlarged due to the rising number of customers and their related features. This study addresses this problem to present pre-results of a credit card customer segmentation for a feature-oriented campaign targeting model by using Principal Component Analysis (PCA). The dataset of 100 credit-card customers from a large private bank in Türkiye with respect to 17 features under the Demographic, Spending Habits, and Past Campaign Participation clusters is used for the solution of the problem. As result, a considerable high correlation between 5 pairs is observed. We believe that these pre-results can be an appropriate initial solution to form a integrated model where decreased number of features can be used for the success criteria of the determination of appropriate credit card campaigns.