BUSINESS ANALYTICS FOR CUSTOMER SEGMENTATION AND MARKETING CAMPAIGN EFFECTIVENESS: A MACHINE LEARNING APPROACH

Abstract

The increasing availability of customer data has transformed marketing decision-making by enabling organizations to adopt business analytics and machine learning for customer-focused strategies. This study investigates customer segmentation and marketing campaign effectiveness using the Customer Personality Analysis dataset comprising 2,240 customer records. A quantitative research design based on secondary data was employed to examine customer demographic characteristics, purchasing behaviour, and campaign responses. Data preprocessing included missing value imputation, outlier treatment, feature encoding, and data standardization. Customer segmentation was performed using the K-means clustering algorithm, while five supervised machine learning models—Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and Extreme Gradient Boosting (XGBoost)—were developed to predict marketing campaign responses. Model performance was evaluated using accuracy, precision, recall, F1-score, ROC-AUC, and confusion matrix analyses. The findings identified four distinct customer segments with significant differences in income, spending behaviour, and campaign engagement. Random Forest achieved the highest predictive performance, demonstrating its effectiveness in identifying customers with a greater likelihood of responding to marketing campaigns. Income, product expenditure, recency, and previous campaign participation emerged as the most influential predictors of campaign response. The proposed business analytics framework provides practical guidance for customer segmentation, personalized marketing, and strategic resource allocation. The study contributes to marketing analytics literature by demonstrating how machine learning can improve campaign effectiveness and support data-driven managerial decision-making in competitive business environments.