<![CDATA[Micro, Small, and Medium Enterprises (MSMEs) in Indragiri Hilir Regency (INHIL) play a strategic role in the local economy, particularly through the marketing of local specialty products such as dodol, crackers, and syrup. However, the majority of MSME operators still face challenges in optimally utilizing historical transaction data to formulate effective and measurable marketing strategies. This study aims to design and implement a web-based e-commerce system integrated with a smart customer segmentation module using a K-Means Clustering algorithm based on the Recency, Frequency, Monetary (RFM) metric. Methodologically, transaction data is extracted from a relational database using PHP (PDO) and MySQL. To ensure clustering accuracy, the data is processed through Min-Max Scaling normalization to eliminate scale bias between high-value nominal variables and low-frequency ones. The algorithm then calculates Euclidean distances to objectively group customers into three distinct segments. The system’s key advantage lies in its dynamic labeling mechanism (intelligent labeling), which automatically names clusters based on customer value scores, resulting in clear categories: “Loyal/Royal,” “Potential,” and “Needs Attention.” System testing results show that this platform successfully processes and visualizes customer data in real-time through an interactive and responsive analytics dashboard. This implementation not only modernizes the user interface but also provides concrete data-driven insights for business managers. Consequently, SME administrators can design personalized promotional strategies, improve customer retention, and significantly optimize revenue, while also serving as a practical reference for accelerating the digital transformation of SMEs in rural areas.]]>
Copyrights © 2026
