Substance use is prevalent among South African adolescents, but few interventions exist to reduce risk of harm. This study assesses the feasibility, acceptability and preliminary effects of a brief intervention for reducing adolescent substance use and other risk behaviours. This single‐arm feasibility test recruited 30 substance‐using adolescents and their primary caregiver. Participants received separate interventions (2 sessions for adolescents, 1 session for caregivers), with a subsample randomly selected for post‐intervention interviews. Feasibility was measured by the proportion of eligible adolescents who were enrolled and retained in the study. Interviews explored acceptability, and changes in outcomes from baseline to 1‐month follow‐up assessed preliminary effects of the intervention. Thirty of 43 (69.8%) eligible adolescents and their caregivers were enrolled, with 29 adolescents (96.7%) and 28 caregivers (93.3%) completing the intervention. Twenty‐eight adolescents (93.3%) and 29 caregivers (96.7%) were retained at follow‐up. Frequency of alcohol, cannabis use and delinquent‐type behaviours decreased significantly from baseline to follow‐up. Participants appreciated the intervention content and delivery and felt that it facilitated behaviour change. Suggestions for improving the intervention were provided. This study found that the intervention is feasible, acceptable and had promising effects on adolescent behaviour. Efficacy must be established with a randomised controlled trial.