We present practical lessons learned from applying the recent close-kin mark–recapture (CKMR) abundance estimation method to thornback ray (Raja clavata). For CKMR, related individuals are identified from their genotypes and their number and pattern is used for abundance estimation. We genotyped over 7000 individuals collected in the Bay of Biscay using Single Nucleotide Polymorphism (SNP) markers finding 99 parent–offspring pairs. The estimated number of adult thornback rays in the central Bay of Biscay was around 135000 (CV 0.19) in 2013. In total, four lessons were drawn: (i) CKMR helps identifying metapopulation structure, which if ignored might affect abundance estimates and/or time trends. There was strong evidence for two distinct local populations of thornback ray with no demographic connectivity. (ii) Demographic sample composition can affect precision and needs to include a range of birth years, which turned out to be difficult for thornback ray. (iii) Reasonable age information for potential offspring is essential. (iv) The sex of potential parents is needed and might be identified from sex-related SNPs. Reliable abundance estimation by CKMR appears feasible for a wide range of species provided that: sampling adequately covers potential local population structure, has appropriate demographic composition, and the age of potential offspring is reasonably well-known.