The idea that context is important when predicting customer behavior has been maintained by scholars in marketing and data mining. However, no systematic study measuring how much the contextual information really matters in building customer models in personalization applications have been done before. In this paper, we address this problem. To this aim, we collected data containing rich contextual information by developing a special-purpose browser to help users to navigate a well-known e-commerce retail portal and purchase products on its site. The experimental results show that context does matter for the case of modeling behavior of individual customers. The granularity of contextual information also matters, and the effect of contextual information gets diluted during the process of aggregating customers? data.