The rapid development of the Internet has made the number of candidate Web services satisfying users’ needs grow rapidly. How to effectively find Web services that meet the needs of users’ Mashup from such a large-scale Web service collection has become a major research issue. Therefore, aiming at how to recommend appropriate Web services to build high-quality Mashup application, this paper proposes a trusted hybrid recommendation method of Web services considering timeliness combining rating timeliness, trust and tag information. First, in the use of rating, the time of service rating is considered to solve the problem of user preference transfer. Secondly, considering the trust relationship between users in the social network, the trust degree between users is calculated to solve the malicious recommendation. Finally, so as to alleviate the issues of data sparsity, the tag information has imported and utilized to improve the recommendation method of Web services. The experimental results show that compared with other existing methods, the method could improve the accuracy and credibility of service recommendation and have a good ability to resist malicious recommendation attacks.