A large proportion of protein-protein interactions are mediated by families of peptide-binding domains. Comprehensive characterization of each of them is critical for understanding the mechanisms and networks of protein interaction at the domain level. However, existing methods are all based on large-scale screenings for each domain, which are inefficient to deal with hundreds of members in major domain families. We developed a systematic strategy for efficient binding property characterization of peptide-binding domains based on high-throughput validation screening of a specialized candidate ligand library using yeast two-hybrid mating array. Its outstanding feature is that the overall efficiency is dramatically improved rather than traditional screening, and it will be higher as the system cycles. PDZ domain family was first used to test the strategy. Five PDZ domains have been rapidly characterized. Broader binding properties have been identified compared to other methods, including novel recognition specificities that provided the basis for major revision of conventional PDZ classification. Several novel interactions have been discovered, serving as significant clues for further functional investigation. This strategy can be easily extended to a variety of peptide-binding domains, as a powerful tool for comprehensive analysis of domain binding property in proteomic scale.