Nutritional issues among toddlers continue to be a pressing public health challenge in Indonesia, including in Kelurahan Pekan Kuala, where although anthropometric data have been systematically collected through the e-PPGBM application, they have not been thoroughly explored in terms of clustering patterns that may provide deeper insights. This study seeks to classify toddler nutritional status by applying the K-Means Clustering method to anthropometric indicators such as age, weight, height, and weight-to-height index. A dataset consisting of 648 entries recorded between January and March 2025 was processed using MATLAB R2014b with cluster variations set at 5, 7, and 9. The analysis revealed that the majority of toddlers were categorized as having good nutritional status, while a portion of the sample was identified as undernourished and some at risk of overnutrition, indicating the diverse nutritional challenges faced by this community. Furthermore, testing the variance across cluster configurations demonstrated that the 9-cluster model yielded the lowest variance score of 0.20, thereby representing the most optimal solution since it produced more homogeneous, balanced, and stable clusters compared to other configurations. These outcomes highlight the importance of data-driven approaches in public health planning, as the clustering results not only provide a clearer picture of nutritional distribution among toddlers but also serve as a foundation for more evidence-based and targeted intervention strategies. By offering a more granular understanding of nutritional variations, this research is expected to support local health authorities in developing customized nutrition programs, allocating resources more effectively, and ultimately improving child health outcomes in Kelurahan Pekan Kuala and similar communities across Indonesia, where malnutrition and overnutrition risks continue to coexist.