This study aims analysis the association between regional culture and Expansion Levels in e-commerce business. This study was designed through an archival approach, and the data has information on 3,065 e-commerce sellers with 100,000 customer purchase orders. We use the regional cultural dimensions of Hofstede et al. (2010) for electronic sellers and the averages of the cultural dimensions of their respective customers. Therefore, to describe the association, we use a machine learning technique, a decision tree, to predict the level of regionalization. The findings described the importance of the long-term orientation dimension. In all decision tree models, this dimension was the most important, as it was decisive for dividing e-commerce sellers into low and high Expansion Levels. Moreover, other dimensions were also relevant. For example, power distance was also highlighted in models that considered the characteristics of customers or sellers only. The sample is only limited to a small business that needs support to develop their sales. The implication of these results is based on how the expansion of e-commerce can be associated with cultural dimensions. And yet, it is noted that the characteristics of the customers were more important to predict the Expansion Level since it obtained the smallest error.