In this study, we aim to investigate the influence of the time spent reading product information on consumer behavior in e-commerce. Given the rapid growth of e-commerce and the increasing importance of understanding online consumer behavior, our research focuses on gaining a deeper understanding of customer navigation on e-commerce websites and its effects on purchasing decisions. Recognizing the multidimensional and dynamic nature of consumer behavior, we utilize machine learning techniques, which offer the capacity to handle complex data structures and reveal hidden patterns within the data, thereby augmenting our comprehension of underlying consumer behavior mechanisms. By analyzing clickstream data using Machine Learning (ML) algorithms, we provide new insights into the internal structure of customer clusters and propose a methodology for analyzing non-linear relationships in datasets. Our results reveal that the time spent reading product-related information, combined with other factors such as bounce rates, exit rates, and customer type, significantly influences a customer’s purchasing decision. This study contributes to the existing literature on e-commerce research and offers practical implications for e-commerce website design and marketing strategies.