This research discusses the application of the K-Means Clustering algorithm to analyze the usage patterns of wartel services at the Salafiyah Syafi'iyah Sukorejo Islamic Boarding School. The purpose of this research is to group users into several clusters based on call duration, frequency of use, and total call cost. User data was analyzed using the stages in the SEMMA method (Sample, Explore, Modify, Model, Assess) to ensure systematic and structured data processing. The results showed that the K-Means algorithm was able to form three main clusters, namely users with low, medium, and high intensity. The majority of users belong to the low-intensity cluster with short average call duration and minimal expenditure, while the high-intensity cluster consists of users who make long calls with high costs. Further analysis shows that the highest usage time is at night (19.00-21.00). Based on these results, it is recommended that wartel managers optimize operating hours and provide promotional call packages according to the needs of each user cluster. In addition, diversification of services such as cheap internet access can also increase the attractiveness of wartel in the digital era. This research uses clustering methods to assist data-based strategic decision-making, as outlined by Han and supported by the application of SEMMA from SAS Institute (1998).