Non-pharmacological interventions for attention-deficit/hyperactivity disorder are useful treatments, but it is unclear how effective school-based interventions are for a range of outcomes and which features of interventions are most effective. This paper systematically reviews randomised controlled trial evidence of the effectiveness of interventions for children with ADHD in school settings. Three methods of synthesis were used to explore the effectiveness of interventions, whether certain types of interventions are more effective than others and which components of interventions lead to effective academic outcomes. Twenty-eight studies (n = 1807) were included in the review. Eight types of interventions were evaluated and a range of different ADHD symptoms, difficulties and school outcomes were assessed across studies. Meta-analyses demonstrated beneficial effects for interventions that combine multiple features (median effect size g = 0.37, interquartile range 0.32, range 0.09-1.13) and suggest some promise for daily report card interventions (median g = 0.62, IQR = 0.25, range 0.13-1.62). Meta-regression analyses did not give a consistent message regarding which types of interventions were more effective than others. Finally, qualitative comparative analysis demonstrated that self-regulation and one-to-one intervention delivery were important components of interventions that were effective for academic outcomes. These two components were not sufficient though; when they appeared with personalisation for individual recipients and delivery in the classroom, or when interventions did not aim to improve child relationships, interventions were effective. This review provides updated information about the effectiveness of non-pharmacological interventions specific to school settings and gives tentative messages about important features of these interventions for academic outcomes.