For long‐lived species, adult survival is the parameter that theoretically has the strongest impact on population dynamics; this makes its estimate crucial for conservation management. The use of non‐invasive genetic approaches coupled with capture–recapture methods is being used increasingly to estimate adult survival. Despite being regularly employed in mammals, it has been little‐used in birds and notably in long‐lived territorial species. This study aimed to determine if feathers obtained from eagle nests could be used to estimate the adult apparent survival rate and the costs associated with such an approach. Feather samples (n = 180) were collected in nests (n = 78) and directly from individual Golden Eagles (n = 36) Aquila chrysaetos in France and were genotyped using 17 microsatellite markers. The genotyping error rates obtained were low, with 3% amplification failures, 3.1% dropouts and 1.7% false alleles. The genetic variability of the markers was high, with a probability of identity between siblings of 6.8 × 10−6, allowing reliable individual identification. Of the feathers collected in nests, 90% were from breeding females; this allowed apparent survival to be estimated for adult females at a reasonable cost. The genotyping of three feathers collected from a nest ensured individual identification of the breeding female with near certainty. Our simulations showed that monitoring of at least 20 pairs over 5 years, or 10 pairs over 10 years is necessary to detect a 10% absolute decrease in adult apparent survival. Estimating the adult survival rate of long‐lived birds using feathers collected in nests is possible. A substantial budget, albeit one that is likely to be lower than that for a common capture–recapture survey, would be needed to obtain precise survival estimates.