Web-based customization is widely adopted in a variety of domains today. Current conceptualization of customization is to provide individualized messages to message recipients based on their particular needs or preferences. A growing body of empirical research has shown positive effects for customization, suggesting that customized messages generate stronger memory and a more favorable attitude than non-customized ones because they match message recipients' need for unique self identity. However, such findings are centered around the Western notion of self which values individuality - prior studies have tested customization effects with only American people. Given that people from different cultures tend to have different views of self, positive customization effects might not be realized in those cultures that do not encourage unique self identity (e.g., East Asian cultures). To advance conceptual understanding of customization, this dissertation tested existing conceptual approaches to customization (tailoring, targeting, and generic) with two groups of participants from different cultures (Americans and Chinese). More specifically, a 3 x 2 full factorial between-subjects main experiment was designed. The customization factor was manipulated by exposing participants to Web sites with tailored, targeted, or generic messages. The culture factor was measured. The main experiment was conducted following a pilot study that examined the effectiveness of customized message manipulation, measures, experiment procedures, and sampling method. It was found that customized messages in general (including both tailored and targeted messages) generated stronger memory and a more favorable attitude than non-customized messages (generic messages). Furthermore, an interaction effect was detected between customization and culture on attitude. Tailored messages generated the most favorable attitude for American participants (individualists) because these messages matched their preferences for unique self identity. In contrast, targeted messages generated the most favorable attitude for Chinese participants (collectivists) since these messages matched their preferences for self-in-group identity. Such customization effects were mediated by three mediating variables: perceived relevance, perceived involvement, and psychological sense of community. Theoretical and practical implications of study findings were discussed. Study limitations and future research directions were also addressed.