Many people overestimate the amount of alcohol that increases their risk of harm and so may not perceive any need to change their drinking behaviour. Several countries have developed low-risk drinking guidelines, yet awareness of these guidelines remains low. Furthermore, mass media campaigns about alcohol-related harms may have limited impact if people do not perceive their current consumption as potentially harmful. Integrating drinking guidelines into media campaigns about alcohol's harms can concurrently provide drinkers with information about low-risk drinking levels and compelling reasons to comply. Our aim was to build understanding of the effectiveness of messages about the long-term harms of drinking and low-risk drinking guidelines, by testing the mediating effects of estimates of harmful drinking levels and attitudes towards drinking alcohol on subsequent intentions and behaviours. In an online experiment conducted in 2016, n = 1156 Australian adult monthly drinkers were randomly assigned to view advertisements for non-alcohol products (NON-ALC; control), advertisements featuring long-term harms of alcohol (LTH), or LTH advertisements plus a guideline message (LTH + G). Immediately following exposure, we measured estimates of harmful drinking levels and attitudes towards drinking alcohol. One week later, we measured intentions to drink less and behavioural compliance with the guideline. Compared to NON-ALC advertisements, exposure to LTH + G advertisements increased (i) the proportion of respondents who correctly estimated harmful drinking levels, which in turn, strengthened intentions to drink less (42% of the total effect was mediated), and (ii) negative attitudes, which in turn, also increased intentions to drink less (35% mediated) and behavioural compliance (24% mediated). Compared to NON-ALC, LTH advertisements increased negative attitudes, which in turn strengthened intentions to drink less (53% mediated). When paired with effective alcohol harm reduction television advertisements, messages promoting low-risk drinking guidelines can increase drinkers’ intentions to reduce their alcohol consumption and compliance with low-risk drinking guidelines. •Uses experimental design with one-week follow-up to examine message effects.•Correct estimates of low-risk drinking levels mediate effects on intentions.•Negative attitudes towards drinking mediate effects on intentions and behaviour.•Low-risk drinking guideline messages should be added to the end of advertisements.