There is an extraordinary assortment of technical approaches to conserving carnivore populations, but the effectiveness of conservation activities is rarely evaluated. Accordingly, we initiated a study to assess the impact of several conservation interventions on the dynamics and persistence of a leopard (Panthera pardus) population in Phinda Private Game Reserve, South Africa. These included revisions of the statutory systems that regulate problem animal control and trophy hunting, and we instituted a program intended to reduce human–leopard conflict in the region. We compared demographic rates of radiocollared leopards from two sampling periods: a pre-treatment period prior to intervention, and a treatment period after intervention. The average annual mortality rate of the population decreased from 0.401 ± 0.070 to 0.134 ± 0.016, with fewer leopards killed by humans and in intraspecific clashes after intervention. The overall reproductive output of the population increased in the treatment period, although annual cub production was higher in the pre-treatment period. This was mainly due to larger litter sizes prior to intervention, which may have been a strategy used by female leopards to offset high levels of infanticide. Results from camera-trap surveys and a Leslie-matrix model indicated an increase in annual population growth rate (λ) of 14–16% after the implementation of conservation measures. Our findings highlight the importance of addressing both the numerical and functional components of population dynamics when managing large carnivores exposed to hunting or persecution. Keywords: Evidence-based conservation, Trophy hunting, Problem animal control, Human–carnivore conflict, Phinda–Mkhuze.