The rapid decline in genotyping costs over the past two decades has resulted in an unprecedented increase in the number of genotyped animals across all livestock species. In contrast, phenotyping complex traits, such as health, behavior, resilience, and efficiency, remains expensive and labor-intensive. Consequently, the limited availability of high-quality phenotypic records has become a major constraint for exploring the genetic architecture of these traits. However, the abundance of genomic data creates new opportunities to strategically select the most informative individuals for phenotyping based on their genotype information, an approach we term GENIUS (Genotype-Informed Unbiased Selection). Using a simulated dataset comprising 2,326 individuals genotyped for 10,031 SNPs and segregating three additive quantitative trait loci (QTL), we evaluated the utility of two strategies for identifying informative individuals for phenotyping: (i) a principal component analysis (PCA) based approach and (ii) a redundancy-reduction (RR) approach aimed at maximizing genetic diversity. The performance of these strategies was compared with random sampling. Both approaches achieved higher QTL detection rates than random selection across reduced subsets of 1,000, 500, and 250 individuals. Notably, the RR strategy outperformed the PCA based method and successfully detected all three QTL using phenotypic information from only 250 individuals, approximately 10% of the dataset. These results demonstrate that GENIUS enables efficient identification of phenotyping candidates and provides a feasible and highly effective strategy to substantially reduce phenotyping costs while maintaining strong power to detect QTL in complex populations.