Who sells knowledge online? An exploratory study of knowledge celebrities in China

Purpose This study explores identity signaling used by an emerging class of knowledge celebrities in China - Knowledge Wanghong - who sell knowledge products on online platforms. Because identity signaling may involve constructing unique online identities and controlling over product-related and seller-related characteristics, the purpose of this study is two-fold: (1) to uncover different online identities of knowledge celebrities; and (2) to examine the extent to which the online identity type is associated with their product-related characteristics, seller-related characteristics and sales performance. Design/methodology/approach A unique data set was collected from a Chinese leading pay-for-knowledge platform - Zhihu - which featured the online profiles of tens of thousands of knowledge celebrities. Online identity types were derived from their self-edited content using Latent Dirichlet Allocation (LDA) topic modeling. Thereafter, their product-related characteristics, seller-related characteristics and respective sales performance were analyzed across different identity types using analysis of variance (ANOVA) and multiple-group linear regression. Findings Knowledge celebrities are clustered into four distinctive online identities: Mentor, Broker, Storyteller and Geek. Product-related characteristics, sell-related characteristics and sales performance varied across four different identities. Additionally, the online identity type moderated the relationships among their product-related characteristics, sell-related characteristics and sales performance. Originality/value As emerging-phenomenon-based research, this study extends related literature by using the notion of identity signaling to analyze a peculiar group of online celebrities who are setting an important trend in the pay-for-knowledge model in China.