A comparison of four different clustering algorithms namely K-means, Hierarchical Clustering, DBSCAN, and Spectral Clustering to identify the best algorithm is implemented in the proposed methodology, which is used to group customers based on two transactional historical data obtained from kaggle. Each algorithm’s performance is evaluated using silhouette scores and execution time. The proposed methodology assesses the quality of the obtained clusters and the time complexity of the analyzed algorithms, pointing out their strengths and weaknesses. The results draw attention to the issue of the selection of accurate clustering methods to get the best customer segmentation strategies and thus denote Spectral Clustering as a technique which outperforms the rest in sense of their silhouette scores, which point towards its better clustering quality. Yet, K-means and DBSCAN are found to be quite powerful in terms of lowering computational costs and identifying noise, respectively. Hierarchical Clustering gives a complete description of the entire hierarchical structure of clusters. These pieces of knowledge are helpful for those organizations in a quest to fine-tune the way they segment their customers in order to enhance marketing efforts and decision making.