This research seeks to understand how communicators’ strategy towards determining if an argument would be perceived as persuasive by someone else may be affected by their social judgements of that person. In so doing, this research contributes to our larger understanding of how social categorization processes affect communication by applying a model of mental state inference to the study of persuasion. Using the framework of the similarity-contingency model, I examine how communicators may change their approach to an argument depending on whether they perceive their message target as similar or dissimilar. It was theorized that when people have more in common with a target, they project, or rely on their own attitudes when determining which arguments would be most likely to persuade their target to support an initiative; when people have less in common with a target, they stereotype, or rely more on what they perceive to be the attitudes of a typical member of their target’s group when determining which arguments would be most likely to persuade their target to support an initiative. I tested this question using two different samples of pro-environmental communicators—activists at a federated national climate advocacy organization and college students at a university in the western United States—to explore whether the model would generalize to explain the communicative decisions of trained communicators and novices in different contexts. In Study 1 (N = 161), I examined if experimentally manipulated similarity of the persuasive target (similar vs. dissimilar) would affect whether environmental activists project or stereotype while ranking the efficacy of various arguments to persuade a local businessman to endorse the Energy Innovation & Carbon Dividend Act. Manipulated similarity was not found to affect whether activists project or stereotype when making choices about which pro-environmental arguments to use on a target. In Study 2 (N = 162), I examined if similarity of the persuasive target (both manipulated and measured via self-report) would affect whether pro-environmental college students in southern California project or stereotype as strategies for choosing which arguments to use to persuade a local businessman to support Carbon Neutrality in Santa Barbara. Although effects were marginal, experimentally manipulated similarity was found to affect students’ approach towards persuasive strategy: students in the similarity condition projected more than those in the dissimilarity condition. Analyses with perceived similarity provided converging support: students who perceived the target of persuasion as similar projected more than students who perceived the target of persuasion as dissimilar. Across both studies, communicators displayed a strong general pattern of stereotyping: collapsing across similarity levels, environmentalists engaged in greater stereotyping than projection when evaluating arguments for a businessman. Demonstrated through the present, the similarity-contingency model offers a new perspective and set of tools to the study of how social judgement affects communication. Further, by studying the mindset and the approach of the communicator instead of the message recipient, this research provides a more holistic view of the variables at play during a persuasive exchange.