The model of developing luminescent materials has transferred from the conventional “trial and error” to “experience- guided experiments”, and further to a novel paradigm of “theoretical prediction and experimental verification”. Efficient theoretical prediction and rapid experimental verification are a key to this transformation. The theoretical prediction methods such as high-throughput calculations and machine learning are becoming more and more mature, and the corresponding experimental methods such as single-particle diagnosis are more efficient, laying a theoretical and experimental foundation for the development of novel luminescent materials. This review briefly summarizes recent research progresses on discovering new rare earth-activated luminescent materials based on single-particle diagnosis and high-throughput computation approaches.