This paper explores the conditions under which one person can persuade another to change their action, using a mechanism design approach. The authors introduce the concept of a "persuasion mechanism," which is a game between a Sender and a Receiver defined by an information structure and a message technology. They derive necessary and sufficient conditions for the existence of a persuasion mechanism that strictly benefits the Sender and characterize the optimal mechanism. The paper also analyzes several examples to illustrate the applicability of their results. Key findings include that Sender benefits from persuasion if and only if there exists a Bayes-plausible distribution of posteriors such that the expected utility of Sender is strictly greater than the utility without persuasion. The optimal mechanism is characterized by the concave closure of Sender's utility function and the distribution of posteriors that maximizes this expected utility. The paper also discusses the implications of these results for various economic contexts, such as advertising, courts, and political campaigns.This paper explores the conditions under which one person can persuade another to change their action, using a mechanism design approach. The authors introduce the concept of a "persuasion mechanism," which is a game between a Sender and a Receiver defined by an information structure and a message technology. They derive necessary and sufficient conditions for the existence of a persuasion mechanism that strictly benefits the Sender and characterize the optimal mechanism. The paper also analyzes several examples to illustrate the applicability of their results. Key findings include that Sender benefits from persuasion if and only if there exists a Bayes-plausible distribution of posteriors such that the expected utility of Sender is strictly greater than the utility without persuasion. The optimal mechanism is characterized by the concave closure of Sender's utility function and the distribution of posteriors that maximizes this expected utility. The paper also discusses the implications of these results for various economic contexts, such as advertising, courts, and political campaigns.