Sales Segmentation Analysis of Tobacco Products Using the K-Means Clustering Method

Sonatha, Yance
Erianda, Aldo
Fitri, Redhatul

INVOTEK: Jurnal Inovasi Vokasional dan Teknologi
Vol 24 No 2 (2024): INVOTEK: Jurnal Inovasi Vokasional dan Teknologi

Technological advancements have encouraged businesses to optimize data utilization, including in sales analysis. This study analyzes sales transaction data of tobacco products at Tobacco Shop Taste using the K-Means Clustering method. By implementing the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework, the sales data were categorized into three groups: highly sold, moderately sold, and less sold. These clustering results support stock management, marketing strategies, and data-driven decision-making. A web-based system was developed, providing real-time monitoring of analysis results, which distinguishes this study from existing solutions by enabling store management to promptly respond to sales trends. This study significantly contributes to the application of data mining technology in the tobacco retail sector, despite being limited to a single store and basic variables. Future development opportunities include integrating broader datasets and analyzing external variables to enhance the accuracy and relevance of the findings.