The persistence of risky drinking among young adults in college calls for continued efforts to prevent alcohol-related harms. This study tested a novel prevention strategy targeting a specific mechanism of change: creating attitude-behavior dissonance. Informed by an extensive literature showing strong and consistent associations between alcohol attitudes and drinking behavior, we adapted a brief counterattitudinal advocacy (CAA) manipulation to the alcohol prevention context. We conducted a small randomized controlled trial with 49 heavy-drinking students who reported drinking 4/5 drinks in 1 sitting (female/male, respectively) and endorsed ≥2 alcohol-related consequences in the previous month. Participants were randomly assigned to 1 of 2 conditions (CAA or control). We examined the number of words written in response to the prompts, evaluated adherence to the core theoretical principles used in CAA, and calculated between- and within-group effect sizes on preliminary outcomes (drinking intentions and alcohol consumption). We found that participants in the CAA condition wrote more, and the experience was perceived as consistent with CAA theory. The between-group effect sizes on drinking intentions and drinks per week were d = .36 and d = .27, respectively, and the within-group effect size for drinks per week was d = .83 for those in the CAA condition. Overall, the results of this pilot study support the feasibility and acceptability of CAA-based alcohol use interventions. This study represents a promising first step in adapting CAA-based interventions for use in the alcohol use domain.