<![CDATA[This study aims to identify the level of adoption of digital technology and analyze user behavior on PesonnaGo's internal e-commerce platform through the use of transaction log data. The research uses an exploratory quantitative approach with the Recency-Frequency-Monetary (RFM) Analysis method which is extended through the Point Usage variable and the K-Means Clustering algorithm. The research data was obtained using purposive sampling techniques on 8,888 successful transactions which were then aggregated into 5,007 unique users during the period of November 2025 to February 2026. Data processing is carried out using Python (scikit-learn) in the Google Colaboratory environment. The number of clusters was determined through the Elbow Method and Silhouette Analysis, while the model stability test was carried out using the Adjusted Rand Index (ARI). The results of the study showed the formation of three groups of users, namely High Adoption, Moderate Adoption, and Low Adoption. The High Adoption group had the highest level of activity as indicated by the lowest recency value, the highest frequency, and the largest monetary value and point usage compared to the other groups. In addition, Pearson's correlation analysis showed a very strong positive relationship between transaction value and point usage (r = 0.947). The findings of the study indicate that the behavioral analytics approach based on transaction log data is able to identify the level of technology adoption objectively and provides a basis for the development of service personalization strategies and optimization of user loyalty programs. Keywords : Digital Technology Adoption; User Behavior Analytics; RFM Analysis; K-Means Clustering; Transaction Log Data.]]>
Copyrights © 2026
