The task of determining the correctness of self-positioning of regions in terms of verifying the compliance of texts of regional development strategies with their industry specialization was solved in the course of the research presented in the article. Using the "smart" benchmarking methodology, as well as the Text Mining tools, long-term development strategies of 11 regions with a total text corpus of 415,780 words were analyzed. The main sections of the all-Russian classifier of economic activities that characterize the sectoral priorities of regional development were selected as keywords. The extraction of key concepts from strategy texts, as well as their quantitative analysis, was carried out using the high-level Python programming language. The obtained quantitative results of comparing the named entities of the development strategies of the subjects of the Russian Federation proved that the insufficiency of unique goal-setting in terms of identifying promising specializations in regional development strategies distorts the system of priority development directions. This is objectively one of the reasons why the territories do not achieve the planned indicators. The paper uses methods of text mining, mathematical statistics, grouping and generalization, as well as techniques for visualizing the analyzed data. The author's method of conducting intellectual analysis of texts is universal for any field of science. The developed algorithms for extracting named entities from the text and algorithms for quantitative analysis of the text open up wide horizons for further research in the field of strategy analysis, as public documents addressed to interested subjects.