Women are underrepresented at all levels of elected office. It is suspected that gender stereotypes hinder the electoral success of female candidates, but empirical evidence is inconclusive on whether stereotypes have a direct effect on voting decisions. This empirical conflict stems, in part, from the assumption that voters automatically rely on gender stereotypes when evaluating female candidates. This study explicitly tests the assumption of automatic stereotype activation. I suggest that stereotype reliance depends on whether stereotypes have been activated during a campaign, and it is only when stereotypes are activated that they influence evaluations of female candidates. These hypotheses are tested with a survey experiment and observational analysis. The results show that campaign communication activates stereotypes when they otherwise might not be activated, thereby diminishing support for female candidates.