Objective: The authors tested a prototype intervention designed to deter alcohol use in residence halls. Participants: Approximately 384 freshmen participated in the study over a 2-year period. Methods: The authors devised a feedback method that assessed residents' blood alcohol concentration (BAC) at night and allowed the readings to be retrieved the next day via the Web. Residents in an intervention hall received their BAC readings as well as normative feedback. In a comparison hall, residents could retrieve only the BAC readings. Results: The authors found statistically significant hall differences, but they were small in size and not meaningful. Conclusions: Qualitative findings suggest the intervention had an overall positive impact, but the actions of a subgroup of rebellious drinkers might have obscured the effect. Social norms interventions could provoke some episodes of excessive drinking in students who find these messages objectionable. More research is needed to evaluate delayed BAC feedback. (Contains 2 tables.)