It is often assumed that interventions aimed at supporting students’ self-regulated learning (SRL) are effective for improving achievement because these interventions support SRL activity. In this study, meta-analytic structural equation modeling (MASEM) was used to test whether SRL activity indeed mediates the effect of SRL interventions on achievement in higher education. Contrary to popular belief, the results only provide evidence for partial mediation. Furthermore, three separate meta-analyses were performed to investigate the role of possible moderators of the relations between: (1) SRL interventions and achievement, (2) SRL interventions and SRL activity, and (3) SRL activity and achievement. Although SRL interventions were effective in improving SRL activity and achievement, most of the study, measurement, and intervention moderators did not explain significant variance of the investigated effect sizes. Other factors, such as task motivation and time on task, potentially influence the effectiveness of SRL interventions. Practical, theoretical and methodological implications are provided. •We review the effect of self-regulated learning (SRL) interventions on achievement.•SRL interventions are successful in improving achievement.•SRL activity only partially mediates the effect of SRL interventions on achievement.•Moderator analyses exploring variance in effectiveness were mostly insignificant.•Practitioners are advised to support learners' self-regulated learning.