Unintentional mortality of endangered carnivores due to non-selective trapping is important for conservation and warrants urgent attention. Currently, non-selective traps are being approved and used based on trap selectivity tests conducted according to International Organization for Standardization (ISO) guidelines. We review these guidelines and find them inadequate, because: (1) the ISO definition of selectivity does not account for relative abundance of target and non-target species and does not therefore meaningfully reflect selectivity; (2) the guidelines methodology at best quantifies relative selectivity of one trap against another, which is of limited use unless the control trap is known to have an acceptable level of absolute selectivity for the target species; (3) information on relative trap selectivity cannot simply be extrapolated elsewhere, unless species assemblage and relative species abundances are consistent. We demonstrate that the ISO definition of trap selectivity is only a simple capture proportion and therefore does not represent trap selectivity. ISO guidelines on trap selectivity should be reviewed to reflect particular ecological scenarios and we suggest how this might be done. Policy-makers, practitioners and researchers should interpret scientific results more cautiously. Trap approval decisions should be based on scientific evidence to avoid undermining the conservation of biodiversity.