Customer Segmentation for Targeted Marketing: Exploring Dbscan & K-Means

Knowing customers and segmenting them is crucial for optimizing marketing strategies. It enablesbusinesses to tailor their campaigns to diverse consumer demographics and behavior. This studyleverages data to address the gap in actionable segmentation frameworks by identifying high-valuecustomer profiles. Preexisting literature emphasizes the usefulness of Recency, Frequency, Mone-tary (RFM) metrics, and K-means clustering in the segmentation of retail customers. Yet there area few studies that integrate demographic insights for a holistic profiling. The goal of this analysisis to segment customers from a marketing campaign dataset using RFM features and demographicattributes to develop targeted marketing strategies. We use a dataset of 2,240 customers (Imakash,2021) for this analysis. The methodology combined RFM feature engineering, K-means clustering(validated via the Elbow Method), DBSCAN clustering, and Principal Component Analysis (PCA)for dimensionality reduction. The analysis successfully identified four diverse clusters: high-valueengaged customers, disengaged users, big spenders, and moderately engaged segments. Incomeand age emerged as key demographic differentiators. We also discuss the practical applications,personalized retention strategies for high-value clusters, and revival strategies for inactive users.This project demonstrates integrating RFM metrics with demographic data to enhance marketingprecision. It also offers actionable insights for improving customer lifetime value and campaign ROI.