Megafires are creating severe conservation problems worldwide for wildlife that have obligate dependencies on plant species that are foundational but fire‐intolerant. Wildfire‐induced loss of native perennials and increases in exotic annual grasses threaten greater sage‐grouse (GRSG, Centrocercus urophasianus) in its sagebrush steppe habitat in western North America. Post‐fire restoration using herbicides, seeding, and planting of native perennials such as sagebrush are common, but there are few assessments of GRSG response to the treatments. We measured the presence of GRSG scat and modeled the probability of GRSG presence (PrGRSG‐scat) in relation to variation in plot‐level and landscape‐level predictors, and land treatments, in an intensive, repeat sampling from 2017 to 2020 of 113,000 ha area burned in 2015 in the Soda Megafire (Oregon and Idaho, U.S.A.). GRSG scat was present in less than 200 of more than 8,000 observations, as would be expected for a philopatric species (i.e. high fidelity to home site) returning to degraded habitat. PrGRSG‐scat was positively associated with sagebrush presence at the plot level and was positively related to elevation, lower‐angle slopes, and proximity to sagebrush seedling outplant islands. The statistical significance of relationships of PrGRSG‐scat to restoration treatments was marginal at best, with the largest effect being a positive response of PrGRSG‐scat to pre‐emergent herbicide sprayed to reduce exotic annual grasses. More time may be required for restored sagebrush steppe to meet GRSG needs or for GRSG to “adopt” the restored vegetation. Moreover, whereas scat is a convenient and non‐invasive method to monitor GRSG, its post‐fire scarcity weakens the strength of statistical inference on GRSG recovery patterns and response to restoration.