Monitoring is a critical component of conservation land management and the choice of methods can influence the final inventory of species recorded. The use of camera trapping has increased in recent years as a cost-effective method to record more species and to identify more cryptic and rare species. In this study we first examined data from detailed field surveys (which did not employ camera trapping) in northern Queensland to examine the abundance and frequency of mammals detected by cage, box and pitfall traps, and spotlighting. We then used data from an additional set of sites that compared these methods with camera trapping. Twenty-five species were recorded in the first data set and 26 species were recorded in the second. Overall, mammals were recorded in exceedingly low numbers, and camera trapping only improved the detection of some species such as larger species that could not be trapped (i.e. dingo Canis familiaris and pig Sus scrofa) or were uncommon (i.e. northern quoll Dasyurus hallucatus and northern brown bandicoot Isoodon macrourus). Our results suggest that survey effort should be substantial and use the most suitable methods to identify management, threat and habitat relationships for potentially precarious wildlife communities.