There is a need for simple and robust techniques for assessment and monitoring of populations of the Endangered snow leopard Panthera uncia to inform the development of action plans for snow leopard conservation. We explored the use of occupancy modelling to evaluate the influence of environmental and anthropogenic features on snow leopard site-use patterns. We conducted a camera trap survey across 480 km2 in Gansu Province, China, and used data from 60 camera traps to estimate probabilities of site use and detection using the single season occupancy model. We assessed the influence of three covariates on site use by snow leopards: elevation, the presence of blue sheep Pseudois nayaur and the presence of human disturbance (distance to roads). We recorded 76 captures of snow leopards over 2,906 trap-days, representing a mean capture success of 2.62 captures per 100 trap-days. Elevation had the strongest influence on site use, with the probability of site use increasing with altitude, whereas the influence of presence of prey and distance to roads was relatively weak. Our findings indicate the need for practical and robust techniques to appraise determinants of site use by snow leopards, especially in the context of the limited resources available for such work.