ABSTRACT Smoking is increasingly concentrated in socioeconomically disadvantaged populations that might not have been reached by the policies and communication campaigns to the same degree as the more prosperous groups. In the rapidly changing media landscape, new approaches to developing and testing messages are needed. This project will use an innovative combination of qualitative, eye tracking, and quantitative (discrete choice experiment, randomized clinical trial) studies to develop and evaluate tobacco education message strategies delivered on digital channels that communicate complex scientific concepts to the public, particularly the priority populations. These strategies will be developed in the context of the FDA-proposed ban on menthol in cigarettes and flavors in cigar products (“flavor ban”). Flavor ban has the potential to save hundreds of thousand of lives, but can be undermined by misinformation and misperceptions. Little research has systematically documented the misperceptions or developed messages to correct them. Our project proposes to address this important gap by developing and testing strategies to reach priority populations with messages on digital channels. This project will provide that research by pursuing these specific aims: (1) Identify effective message attributes for digital channels using a co-creation approach; (2) Quantify the relative importance of different attributes of digital messages using a combination of a discrete choice experiment and an eye- tracking study; and (3) Test the impact of messages on behavioral outcomes in a context of digital social media in a randomized clinical trial with a nationally representative sample of flavored combusted tobacco product users. The results will help eliminate tobacco-related disparities and increase the effectiveness of tobacco educational messages among the population for whom anti-tobacco messages have not been optimally effective.