Conservation and management of brown hyaenas (Hyaena brunnea) is hampered by a lack of information on abundance and distribution, which is difficult and labour-intensive to obtain. However, occupancy surveys offer a potentially efficient and robust means of assessing brown hyaena populations. We evaluate the efficacy of camera trapping for estimating brown hyaena occupancy, and the effect of environmental variables and lures on detection probability. We estimated population density in Pilanesberg National Park, South Africa, at 2.8/100 km², occupancy at 1.0 and model-averaged detection probability at 0.1. Using a fish lure increased detection probability to 0.2 and significantly increased encounter rates. We also found that brown hyaenas are more likely to be detected in areas of scrub or woodland rather than grassland. Our results suggest that 13 camera sites would be needed to achieve an occupancy estimate with S.E. of 0.05, and a minimum of 16–34 sampling occasions (with and without the fish lure) should be used in comparable study areas. We conclude that camera trapping is a viable method of estimating brown hyaena occupancy at local and landscape scales and capture—recapture analysis is also possible at a local scale.