Background: Incentives have been proposed as a method to improve attendance in adult literacy classes. In the UK, several areas have piloted the use of incentives to promote attendance at adult literacy classes. To date no rigorous evaluation of this policy has been undertaken. This paper describes (as far as we are aware) the "only" UK-based randomised controlled trial to evaluate the use of financial incentives in order to promote attendance in classes for adult learners. Methods: We used a cluster-randomised design. Twenty-nine adult literacy classes were randomised in two groups using minimisation. Intervention group learners received 5 British Pounds (US$10) for each class attended. The main outcome was class attendance; the secondary outcome was literacy scores. Results: After allocation, one class was found to be ineligible for the study. In the 28 remaining classes there was a statistically significant reduction of about 1.5 sessions (95% confidence interval (CI) 0.28, 2.79; p = 0.019) attended by the intervention group compared with control, after adjusting for cluster size and baseline scores. The difference in reading scores between the intervention and control group, conditioned on baseline scores, was -2.38 (with controls scoring higher than the intervention group), but this difference was not statistically significant (95% CI-7.40 to 2.57, p = 0.33). Conclusion: Payments to attend adult literacy classes had an adverse effect on attendance. This trial needs urgent replication, ideally with a larger incentive, before this approach is widely used by policy makers. (Contains 2 tables.)