Assessment of socio-economic development of both separate territories and countries as a whole, comparison of the received estimations to identify the development disproportions, lacks, and gaps, application of the best practices to decide the revealed problems promotes the development of balanced state regional development policy. Quite common tasks of such assessment are the ranking of regions and their grouping according to some generalized socio-economic development characteristics. The solution to these problems is complicated by the use of a set of indicators of non-numerical nature because traditional statistical methods are unsuitable for their processing. The paper considers the solution to the problem of constructing a consensus rank based on Kemeny's median. To solve the problem of grouping regions, an approach is proposed based on the picture of regions in the space of partial consensus rankings for economic and social development separately, with the subsequent detection of object stratification in this space. The practical implementation of the proposed approach according to partial rankings is given. A heuristic algorithm was used to construct Kemeny's median. The obtained grouping of the regions of Ukraine reproduced as a whole their order following the resulting consensus ranking. Comparison of the obtained results with the known results of solving such tasks, calculated on the set of metric indicators also showed their consistency.