Although some taxa are increasing in number due to active management and predator control, the overall number of kiwi (Apteryx spp.) is declining. Kiwi are cryptic and rare, meaning current monitoring tools, such as call counts, radio telemetry, and surveys using detection dogs are labor‐intensive, yield small datasets, and require substantial resources or provide inaccurate estimates of population sizes. A noninvasive genetic approach could help the conservation effort. We optimized a panel of 23 genetic markers (22 autosomal microsatellite loci and an allosomal marker) to discriminate between all species of kiwi and major lineages within species, while simultaneously determining sex. Markers successfully amplified from both fecal and shed feather DNA samples collected in captivity. We found that DNA extraction was more efficient from shed feathers, but DNA quality was greater with feces, although this was sampling dependent. Our microsatellite panel was able to distinguish between contemporary kiwi populations and lineages and provided PI values in the range of 4.3 × 10−5 to 2.0 × 10−19, which in some cases were sufficient for individualization and mark–recapture studies. As such, we have tested a wide‐reaching, noninvasive molecular approach that will improve conservation management by providing better parameter estimates associated with population ecology and demographics such as abundance, growth rates, and genetic diversity.