Molecular markers are a useful tool allowing conservation and population managers to shed light on genetic processes affecting threatened populations. However, as technological advancements in molecular techniques continue to evolve, conservationists are frequently faced with new genetic markers, each with nuanced variation in their characteristics as well as advantages and disadvantages for informing various questions. We used a well‐studied population of Tasmanian devils (Sarcophilus harrisii) from Maria Island, Tasmania, to illustrate the issues associated with combining multiple genetic data sets and to help answer a question posed by many population managers: which data set will provide the most precise and accurate estimates of the population processes we are trying to measure? We analysed individual heterozygosity (as internal relatedness, IR) of 96 individuals, calculated using four genetic marker types (putatively neutral microsatellites, major histocompatibility complex‐linked microsatellites, reduced representation sequencing, and candidate region resequencing). We found no correlation in IR values across marker types, suggesting that various genetic markers reflect different aspects of genomic diversity. In addition, some marker types were more informative than others for conservation decision‐making. Reduced representation sequencing provided the highest precision (lowest error) for estimating population‐level genetic diversity, and most closely reflected genome‐wide heterozygosity both theoretically and empirically. Within the conservation context, our results highlight important considerations when choosing a molecular technique for wildlife genetics.