Objectives While there is strong evidence that regular participation in physical activity (PA) brings numerous health benefits to older adults, and interventions to effectively promote PA are being developed and tested, the characteristics and components of the most effective interventions remain unclear. This systematically conducted review of systematic reviews evaluated the effects and characteristics of PA promotion interventions aimed at community dwelling people over 50 years old. Methods Major databases were searched for reviews from January 1990 to May 2015. TIDieR guidelines aided data extraction and the ROBIS tool was used to assess the risk of bias. Primary outcomes were objective and self-reported levels of PA. Indicators of psychological wellbeing and participation rates were secondary outcomes. Results Of 1284 records identified, 19 reviews met inclusion criteria and eight included meta-analyses. Interventions typically incorporated behaviour change techniques (BCTs) and were delivered as face-to-face, remote, group, individual or as combined interventions. Despite their heterogeneity, interventions often resulted in sustained improvements in PA over the study period, typically at 12 months, and led to improvements in general wellbeing. However, ways to ensure effective maintenance beyond one year are unclear. Certain intervention components were more clearly associated with positive effects (e.g. tailoring promotion strategy with combination of cognitive and behavioural elements, low to moderate intensity activity recommended). We found no evidence that certain other intervention characteristics were superior in achieving positive outcomes (e.g. mode of delivery, setting, professional background of the intervention provider, type of PA recommended). Conclusion The evidence suggests that interventions to promote PA among older adults are generally effective but there is uncertainty around the most beneficial intervention components. There are indications that purely cognitive strategies and BCTs might be less suitable for older adults than motivators more meaningful to them, including social and environmental support, and enjoyment coming from being physically active. A whole system-oriented approach is required that is tailored to meet the needs of older adults and aligned with social, individual and environmental factors.