Public Health Significance Statement College students' alcohol use is closely related to perceptions of friends' approval, such that those who think their friends are more approving tend to engage in greater alcohol use. The current study revealed that students' perceptions of friends' alcohol use fluctuated substantially across a single year (four timepoints across 12 months) and students self-reported heavier alcohol use at timepoints that they perceived their friends to be relatively more approving of alcohol use. These findings add novel understanding regarding the nature of normative influences and hold important implications pertaining to the adaptation of norms-based harm-reduction strategies. Objective: Perceptions of friends' approval of drinking behaviors (i.e., injunctive drinking norms) play a central role in shaping college students' alcohol use behaviors. However, we know little about the extent that students' perceptions of friends' approval fluctuate over time and whether there are within-person associations between these injunctive norms and alcohol use. To fill this knowledge gap, we estimated within-person variability in perceptions of friends' approval of alcohol use across a 12-month period and examined within-person associations between perceptions of friends' approval and 3 discrete drinking behaviors: number of weekly drinks, hazardous drinking behaviors, and peak estimated blood-alcohol content (peak-eBAC). Method: A sample of college students (N = 433, 54.82% female, M-age = 20.06) reported perceptions of friends' approval of alcohol use and indices of alcohol use behavior at 4 timepoints across a single year. Results: Descriptive estimates of within-person variability of perceived friends' approval revealed that these perceptions fluctuated considerably across the 4 timepoints. After accounting for between-person effects, longitudinal multilevel modeling revealed significant within-person associations between perceptions of friends' approval and (a) number of weekly drinks, (b) hazardous drinking behaviors, and (c) peak-eBAC levels. Students reported heavier alcohol use at timepoints when they perceived their friends as being more approving than usual. Conclusions: Alongside advancing theoretical understanding of social influences on students' alcohol use, the current findings hold important clinical implications for norms-based harm-reduction strategies. To optimize interventions, norms-based approaches may need to be adaptive over time (e.g., boosters) to map onto within-person fluctuations in perceived injunctive norms.