This paper systematically analyzes the models and processes related to wordof- mouth spreading in social networks. This paper simulates the characteristics and rules of word-of-mouth spreading on social network platforms, adopts network evolution models as well as virus spreading models which can precisely reflect the process of word-of-mouth spreading. By computer simulation, the effect of several kinds of parameters in networks and in wordof- mouth spreading model is analyzed. What has been proved, through parameter analysis, is that the secondary “push” of the key node (opinion leader) in social networks has played a significant role in promoting word-of-mouth spreading. In practical applications, shopkeepers can act appropriately to the situation, which means they put in a second period of advertise appropriately after placing one advertisement at random in order to save costs and increase efficiency.