BackgroundPharmacist counselling is an important service that has been associated with improved outcomes. The primary aim of this review was to identify, describe, and determine the effectiveness of interventions for improving the counselling practice of community pharmacists.MethodsWe searched PubMed (from January 1990 to June 2017) and the Cochrane Library (June 2017). To supplement our database searches, we searched Google Scholar for papers that cited the identified studies. We included only studies that reported the impact of the intervention on pharmacists’ behaviour during counselling. We searched for data from studies with randomised trials, non-randomised trials, controlled before-after studies, or interrupted time series study designs. Parameters including selection bias, performance bias, detection bias, and attrition bias were assessed. The data were narratively synthesised.ResultsWe screened 2335 abstracts and 59 full-text articles and included 17 RCTs. Overall, three studies were determined to have a high risk of bias, and 14 studies were determined to have an unclear risk of bias. Fifteen studies investigated multifaceted interventions that included two or more components. The most commonly used interventions were educational meetings (n = 14), educational materials (n = 9), educational outreach visits (n = 5), feedback (n = 5), guidelines (n = 5), and local opinion leaders (n = 2). Outcomes were measured using simulated patient visits (n = 10), and the self-reported outcomes of patient or pharmacists (n = 6). Most of the included studies (n = 11) reported some degree of improvement in counselling practices.ConclusionsThe included studies showed that educational meetings combined with educational materials, outreach visits, and feedback can improve pharmacist counselling in community settings. However, the unclear risk of bias and poor quality of reporting intervention components necessitate caution in interpreting the findings. Recommendations for future studies based on the evidence gap identified in this review are presented.Electronic supplementary materialThe online version of this article (10.1186/s13643-018-0727-4) contains supplementary material, which is available to authorized users.