Predicting Sales E-Mail Responders using a Natural Language Model

Email is a standard and popular means of establishing potential business relationships between salespeople and future customers, but it is difficult for machines to generate messages that are as convincing as a human author. We take first steps towards the automatic generation of human-quality emails by presenting a model predicting if an e-mail will be successful in eliciting a response. In this paper we propose a natural language model that predicts whether a human-authored sales e-mail will get a response from a previously uncontacted recipient. We test our algorithm with a set of 116 outbound sales e-mails used in practice. Our algorithm is successfully able to predict if an e-mail in this set received a response with an F1 score of 0.81. This work provides initial steps in understanding how to automate convincing communication over email between humans and computers.