Combinatorial chemistry has been proven effective in the search for novel functional materials, especially in the field of organic chemistry, and is being used to identify functional inorganic compounds. However, there is a growing need for approaches that predict and experimentally realize new materials, beyond composition optimization of known systems. Application of combinatorial chemistry to materials discovery is typically hindered by a limited ability to search a wide chemical composition space, and by our ability to experimentally screen promising compounds. Here, a combinatorial scheme is proposed that combines a materials informatics technique to define a chemical search space with highthroughput synthesis and evaluation. We identify high-performance superionic conductors in the Ca-(Nb,Ta)-Bi-O system, demonstrating the effectiveness of this approach for accelerated materials discovery.