The introduction of Web 2.0 and Web 3.0 has changed not only the available web technologies, but also the ways in which users interact. The growing ubiquity of Internet access and a variety of mobile devices have allowed people to choose the most attractive tools and services. The conditions of the created environment are well suited for conducting group decision making processes: a large number of users participate in the network, each of whom has his own interests, knowledge and experience. Despite the huge technological leap, there are still problems to be solved. First, in social networks, people communicate and express opinions with the help of words, while traditional methods of group decision making operate with exact numbers. Experts are required to provide estimates in terms of qualitative aspects. Secondly, it is not enough to find a joint solution for all experts, it is also necessary to reach an acceptable level of consensus. The purpose of this work is to conduct a numerical analysis of the group decision making process in social networks, using user publications as ratings. The paper also proposes a format for conducting a process of reaching consensus and its analysis, using the advantages and features of social networks. © 2019, Springer Nature Switzerland AG.