Excessive alcohol consumption in university students in the UK has been identified by Government agencies and health care providers as significant problem. Social norms research suggests students evaluate and regulate their own alcohol consumption through social comparison with peers. However, students are prone to misperceptions, inaccurately estimating others’ alcohol consumption to be higher than their own. The over estimation of normative peer alcohol consumption has been associated with higher personal alcohol consumption. Unitcheck is an online alcohol resource available to university students in the UK (www.unitcheck.co.uk). Using a social norms approach this online alcohol intervention provides instant personalised normative feedback to students, directly comparing reported individual alcohol consumption with normative peer alcohol consumption. This comparison is designed to correct students’ overestimation of normative peer drinking and encourage moderate alcohol consumption. Currently, Unitcheck feedback is labelled as comparing personal alcohol consumption to normative alcohol consumption of the typical University of Leeds student. There is a paucity of research exploring student responses to personalised normative feedback on an individual level. The aim of this study was to evaluate how students reporting alcohol consumption above recommended weekly limits at the University of Leeds, respond to instant, online personalised normative feedback presented as part of a study version of Unitcheck. A between subjects experimental design employing qualitative think-aloud methodology, plus a semi-structured interview was used. Participants were 21 Undergraduate students in their first or second year of study at the University of Leeds, 67% female, mean age 19.3 years (range 18-21), meanalcohol consumption over the previous week 38.4 units (range 10-150). Participants were randomised to one of two study groups and asked to work through the Unitcheck resource whilst thinking their thoughts aloud. Group A (n=11) used a same-sex referent group in the personalised normative feedback for University of Leeds students. Group B (n=10) used typical University of Leeds student as the normative referent group. All participants then completed a semi-structured interview assessing how believable and personally relevant they found the feedback. Findings from thematic analysis of the transcripts suggest that personalised feedback prompts participants to actively consider their personal value judgements regarding acceptable drinking behaviour. Students responded directly to the normative feedback component. Justifications of ratings of believability of normative feedback were based on personal observations of student drinking behaviour and perceived credibility of the normative data. The current manipulation focussed on gender as a salient comparison group, participants suggested year of study and age as alternative salient normative comparison groups.