The breed conformity evaluation is crucial for the preservation of the traits that characterise each dog breed. The use of genetic markers for this purpose provides a precision and objectivity that can surpass the reliability of phenotypic evaluations. In this study, we present a new simple algorithm for assessing breed conformity. The algorithm creates a similarity matrix based on genotypic data and then uses the median to calculate the percentage of genetic similarity that an individual has in relation to the genetic diversity of the breed. To validate the proposed algorithm, we applied it to the genotypic data of 18 microsatellites and 43,691 single nucleotide polymorphisms (SNPs) of the Kazakh Tazy dog, a breed of great cultural and historical importance to Kazakhstan that is now threatened with extinction due to crossbreeding. The algorithm showed a moderate correlation between the microsatellite and SNP genotyping methods, reflecting the different aspects of genetic similarity. In particular, the SNP-based evaluations agreed better with the expert judgements, highlighting their potential for accurate analysis of breed conformity. The proposed algorithm provides easily interpretable results, is flexible, adapts to different genetic markers and may provide an evaluation mechanism for breed conformity in situations where there is no reference population, incomplete pedigrees, unidentified meta-founders and high genetic diversity in the population.Author summaryOur research was initiated by the urgent concern for the possible extinction of the Kazakh Tazy dog, a breed with deep historical roots and cultural significance in Kazakhstan. Information at the DNA level may lead to faster genetic improvement of the breed than relying only on phenotypic data and pedigrees. Genotypic data can be processed to provide valuable insights into genetic diversity, relatedness and ancestry. However, existing methods do not provide a measure of percentage similarity that can be used to assess how closely a particular individual matches the typical genetic composition of the breed. These challenges have led us to propose an approach that overcomes the limitations of existing methods and allows genetic similarity to be assessed based on genotypes. It is based on a median-based approach to analyze genetic data and has been applied to microsatellite and SNP markers but can also be adapted to other genetic markers. This method provides results even in the absence of a defined reference population, complete pedigrees or identified meta-pedigrees and during high genetic diversity within a breed. It can provide breeders and researchers with a tool to maintain the genetic purity of unique breeds.