The well-known inaccuracy of assessments of the contribution of scientists, organizations, and countries, as well as the half-life of publications and the number of articles and their citations by sections of science when using indexing and citation platforms (using the example of Web of Science) is discussed. This inaccuracy is explained by the fact, that on these platforms, journals are classified by thematic category, while the objects of calculations require classification by articles. The distribution of articles on specific topics among journals is uneven and obeys the Bradford dispersion law, which is not taken into account. It turns out that it is impossible to correct the results of such calculations. The way out is seen in the use of abstract journals for them, in which articles are indexed by headings.