Email Volume Optimization at LinkedIn
Rupesh Gupta*, LinkedIn; Xiaoyu Chen, ; Guanfeng Liang, ; Romer Rosales, LinkedIn; Hsiao-Ping Tseng, ; Ravi Kiran Holur Vijay,
Online social networking services distribute various types of messages geared towards providing increased value to their members. Common types of messages include news, con- nection requests, membership noti_cations, promotions, and event noti_cations. Such communication, if used judiciously, can provide an enormous value to the members. However sending a message for every instance of news, connection re- quest, or the like can result in an overwhelming number of messages in a member’s mailbox. This may result in reduced e_ectiveness of communication if the messages are not su_- ciently relevant to the member’s interests, and potentially a poor brand perception. In this paper, we discuss our strat- egy and experience with regard to the problem of email vol- ume optimization at LinkedIn. In particular, we present a cost-bene_t analysis of sending emails, the key factors to administer an e_ective volume optimization, our algorithm for volume optimization, the architecture of the supporting system, and experimental results from online A/B tests.
Filed under: Optimization Techniques