Web-Based MSME Sales Classification Analysis Information System in Tanjung Rejo Village Using K-Means Clustering and Data Visualization

Thaqif Andika Putra (Muhammadiyah University of North Sumatra Medan)

Ambidextrous : Journal of Innovation, Efficiency and Technology in Organization
Vol. 2 No. 02: Driving Efficiency and Innovation: Transforming Organizations in the Era of Digital A

The development of digital technology has triggered a Big Data explosion, but MSMEs in Tanjung Rejo Village still rely on manual recording, hampering sales performance analysis. This study aims to build a web information system based on K-Means clustering for MSME performance classification using five variables: monthly turnover, transactions, operating hours, strategic location, and operating days. A descriptive quantitative approach was applied to a local MSME population with a purposive sample of 6 representative businesses. The questionnaire instrument collected primary data, analyzed via Min-Max preprocessing, K-Means iteration (k=3), and Chart.js visualization on Python-Flask-SQLite. The results show accurate segmentation into three clusters: undeveloped (2 MSMEs), less developed (3 MSMEs), and developed (1 MSME), with a stable dashboard and successful black-box validation. Conclusion: an effective system supports targeted coaching, practical implications increase MSME competitiveness via data-driven insights, despite limited scalability.