The rapid growth of e-commerce in Indonesia, particularly on the Tokopedia platform, has generated a large volume of customer review data that can be utilized to support business decision-making. This study aims to develop product segmentation based on customer review characteristics using data mining techniques to support Business Intelligence in e-commerce marketplaces. The dataset used is the Tokopedia Product Reviews 2025 dataset from Kaggle, consisting of 5,521 unique products aggregated from the original review data. The study follows the CRISP-DM methodology, including Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. Feature engineering was performed to generate analytical attributes, and the K-Means clustering algorithm was applied with the optimal number of clusters (k = 3), determined using the Elbow Method and Silhouette Score. The clustering results identified three product segments: High Quality (2,749 products), High Demand (one outlier product with exceptionally high sales), and High Volume (2,771 products). The resulting dataset was implemented as a Business Intelligence-ready dataset to support product performance monitoring and data-driven marketing strategy development.