AbstractEstimating abundance is an essential component of monitoring and recovery of rare species, and spatially explicit capture–recapture (SECR) models provide the means for robust density estimation. Previous work has elucidated principles ofSECRstudy design for large, generalist carnivores, but less attention has been paid to study design considerations for smaller species, with less extensive home ranges. Here, we integrated data from an intensive pilot study with simulation modeling to evaluate the influence of survey sampling intensity on precision and accuracy inSECRabundance estimation for a rare lagomorph that specializes on patchily distributed early‐successional habitats. Doing so, we obtained the first mark–recapture density estimates for the New England cottontail (Sylvilagus transitionalis). Capture probability and density on the landscape both impacted the required intensity of the sampling design. The optimal study design for robust estimation also required a greater number of traps relative to home range size or spatial extent than those recommended in priorSECRstudies. This divergence emphasizes thatSECRstudy design considerations will differ among organisms with varying spatial extent and habitat use. Demonstrating the appropriate sampling design for a study system is important prior to embarking in aSECRstudy. Integrating pilot empirical data with simulations provides a powerful means for optimizingSECRstudy design and for facilitating applicability ofSECRapproaches to a wider array of organisms with varying habitat and space use. This methodology may be employed in planning a monitoring program that maximizes effectiveness while minimizing cost and effort, as part of the adaptive management approach to monitor and recover rare or endangered species.