The dynamic evolution of marketing, fueled by advanced data analysis, has significantly improved our capacity to adapt strategies and deepen our understanding of customer behaviors and needs. Customer segmentation is a technique that has been utilized to devise unique strategies for different client types. Through clustering, we aim to identify locations for new shops to achieve comparable outcomes based on a specific revenue target. In our approach, we employ deep embedding clustering (DEC) to simultaneously address cluster assignment and underlying feature representation, while progressively enhancing both the cluster and its feature representation. The suggested implementation enhances the separability between the evaluated clusters, using the Silhouette coefficient as a quantitative comparison criterion.