Details of kills made by lions on Shamwari Private Game Reserve, South Africa, were routinely recorded by staff for three years after reintroduction of the lions, and we used these data to establish the prey profile, prey selection and daily intake rates (kg/FEQ/day). The opportunistic nature of the observations of kills resulted in gaps in the records which we attempted to resolve by analysing both the complete data set and a subset of data in which we omitted kills that were made more than five days apart. The full data set (n = 360 kills) comprised 23 species over three years (1095 days) while the subset (n = 227 kills) comprised 16 species in 368 days. Prey preference (Jacobs' index) was calculated using both data sets and aerial game counts for species availability. Aerial game counts were used both with and without correction for differences in visibility. Prey profiles were very similar for the two data sets with the same species preferred (black wildebeest, ostrich, warthog) and avoided (springbok, impala, common duiker). Prey preferences were calculated separately for each year, and revealed a switch from a preference for blesbok to avoidance, and the reverse for warthog. There was no significant difference in the mean prey size using either data set (full data set,131.5 kg; subset of data,133.1 kg). The two data sets did generate very different daily intake rates, however, with higher values from the subset of data (9.3 kg/FEQ/day year 1) than the full data set (4.6 kg/FEQ/day in year 1). We conclude that kills located by drive vehicles on small ecotourism reserves can be used to establish prey profiles. Removing gaps in the kill lists will increase the accuracy of the profiles and is essential for calculation of kill rates and daily food consumption.