How do personality traits condition the effects of campaign messages meant to mobilize voters? With two nationally representative US survey experiments, I show that common aggressive metaphors mobilize or demobilize voters depending on their traits. Aggressive metaphors increase the mobilizing impact of motivations to participate among aggressive individuals but decrease that impact among low-aggression people. For example, the language mobilizes strong partisans with aggressive personalities but demobilizes strong partisans low in aggression. This heterogeneity showcases the nuanced power of metaphors in campaigns, reaffirms the importance of personality in political behavior, and reveals the hidden role of aggression in non-violent political behavior for the first time. In practice, the net effects of aggressive metaphors can be positive, negative, or null depending on average traits in an audience.