With the application of self-services, companies can keep down costs and customers can enjoy convenient services. However, the market penetration ratio of many self-services is not high. Thus, figuring out the market diffusion law of self-services is significant to the development of self-services. A lot of researches have proved that the interpersonal relationship network has an influence on self-service diffusion. ,However, none of them analyzed the influence of individuals in communities. Therefore, on the base of the Shaikh’s mixture diffusion model, this paper analyzes the influence caused by the number of individuals in different social groups and the number of adopted individuals and builds an HOGI model and an HEGI model. This paper also takes email service and instant messenger service for examples and gives a comparative analysis of the two models by means of econometric analyses. The result shows that under community pressures, the HOGI model has a better fitting efficiency. However, taking the mutual functions between individuals into consideration, the HEGI model has a better fitting efficiency.