An online community is a virtual network where people with common interests and characteristics gather to share information, opinions, and experiences. It plays an important role in the decision-making process of consumers, and its importance has been increasing in firm’s marketing activities. Accordingly, research on the new product diffusion through bandwagon consumption among members within the online community has been drawing attention. This study pays attention to the fact that the social influence of hubs in online communities can vary depending on the preference skewness within the group. We collected community information and users’ purchase data from Last.fm, a world-leading music social network website, and analyzed using a fixed-effect linear regression model. This study derived the following results. First, social contagion on new music purchases was greater in communities with variously distributed preferences than in communities with skewed preferences. Also, hubs had a greater influence than others on the purchase of new products by other consumers. In particular, the positive influence of hubs on the purchase behavior of other consumers increased in the group with skewed preferences, while decreased in the group with diverse preferences. Based on these results, this study has the theoretical implications of presenting a new boundary condition, in-group preference skewness, for the impact of network hubs on new product diffusion. Furthermore, this study offers managerial implications to new product marketers and community managers by providing efficient marketing strategies.