In many social networks, individuals’ actions depend on the actions of their peers and their information. How can a designer design informative signals to induce a desired outcome in a networked system? Candogan answers this question in a setting where the designer is restricted to using public signaling mechanisms. He provides a convex programming framework for obtaining optimal public signaling mechanisms in networks where agents’ actions exhibit local strategic complementarities and characterizes the (double-interval) structure of these mechanisms. The framework he develops is useful in other settings (where the designer’s payoff is an increasing step function of the posterior mean). He also provides an approach for designing asymptotically optimal public signaling mechanisms in large random networks that relies only on using the (limiting) degree distribution information. Finally, he sheds light on another fundamental question: Which networks are more amenable to persuasion?