Although Kazakhstan increases the area under soybean cultivation annually, its production remains challenging in the northern and eastern regions due to the limited number of varieties adapted to climatic conditions. Traditional methods of selecting parental forms for the creation of new ultraearly and early soybean varieties require significant time and resource investments, so modern breeding technologies are aimed at increasing the accuracy and efficiency of hybridization. In the context of agricultural intensification and the need to adapt soybean varieties to specific agroecological conditions, cluster analysis has proven to be an effective tool for evaluating and selecting varietal material. Its application made it possible not only to structure the studied soybean accessions by economically important traits, but also to identify the most promising forms for subsequent use in breeding programs. The article presents the results of a study of 102 collection soybean accessions of various ecological and geographical origins, evaluated by the main economically important traits in order to define criteria for selecting sources and donors for breeding for high productivity, early maturity, and high biochemical performance, using cluster analysis via Ward’s method with the Statistica v.13 software. To conduct a cluster analysis of 102 collection soybean accessions for systematization based on economically important traits and to identify genetically promising parental forms to improve the efficiency of the breeding process and accelerate the development of new varieties adapted to the conditions of the northern and eastern regions of Kazakhstan. The study was conducted using the hierarchical clustering method based on Ward’s method. Statistical analysis was carried out using Statistica v.13. As a result of studying 102 collection soybean accessions using cluster analysis, five clusters were identified that differed in sets of economically important traits. From the first cluster, four accessions were identified as sources and donors of high yield and high protein content in seeds. From the second cluster, six accessions were identified as sources and donors of early maturity, and five as sources of high protein content. From the third cluster, four early maturing accessions, three highyielding accessions, five with high protein content, and three with high fat content were identified. From the fourth cluster, three accessions were identified by yield level, two by 1000 seed weight, two by protein content, and four by pod insertion height. In the fifth cluster, seven soybean accessions stood out for yield and three for pod insertion height. Cluster analysis has proven its effectiveness as a digital breeding tool, facilitating the accelerated development of adapted soybean varieties and the expansion of soybean cultivation in Kazakhstan.