<p>Participation in online gambling is growing and the risk of experiencing harms is also increasing. Brief personalised feedback interventions have been shown to prevent, reduce and address gambling harm and this randomised controlled trial tested the effectiveness of a version customised for the UK. A sample of 1586 online gambling participants with moderate or problem gambling were rapidly recruited from an existing Internet panel of UK residents. Participants were randomised to a no intervention control group or received the self-directed, online intervention which included normative feedback and personalised information explaining the consequences of gambling above lower-risk guidelines. One- and three-month follow-ups assessed the short-term impact on frequency and harm. Feedback and recommendations were collected to guide improvements and increase future utility. All gambling outcomes showed improvement between the initial survey and both follow-ups, however, there were no differences between the intervention and control groups. Most participants displayed normative misperceptions when estimating how much others the same age and gender gambled. The majority of the sample had never previously sought treatment despite over a third of these reporting moderate or problematic levels of gambling. There is need for a publicly available, low-cost alternative to traditional treatment in order to help the large proportion of people with gambling concerns who would not otherwise seek formal support. Although an intervention effect was not detected in this sample, Internet-based alternatives remain a promising opportunity meriting further research.</p>