Customer segmentation is an important step in designing an effective marketing strategy, especially in e-commerce businesses that have a large customer base with diverse shopping behavior characteristics. This study aims to segment customers based on the RFM (Recency, Frequency, and Monetary) approach to identify customer behavior and support more targeted marketing decision-making. The algorithm used in the clustering process is DBSCAN (Density-Based Spatial Clustering of Applications with Noise) because of its ability to find clusters with arbitrary shapes and detect outliers or inactive customers. The research process began with the selection of the dataset, followed by data preprocessing through the stages of cleaning, reduction, RFM transformation, and normalization using the Min-Max method. This study proves that the combination of the RFM approach and the DBSCAN algorithm is effective in segmenting e-commerce customers. The results of this segmentation can be used to develop more personalized marketing strategies, such as active customer retargeting and inactive customer reactivation, so that it can increase marketing efficiency and customer loyalty.
Irawan et al. (Thu,) studied this question.