This paper reviews the main methods for analyzing laboratory data on the 87Sr/86Sr ratio, including the basis interval method and the main variability mapping methods, and analyzes the limitations of their use. The basis interval method involves calculating the mean and two-sigma interval based on samples from one geological area. The weakness of the method is the need to obtain a large sample and the inability of studying variability. The domain mapping method involves plotting mean values for a zone on a color-coded map. The Voronoi diagram limits the zones of different values and demonstrates the heterogeneity of the sample. Geostatistical interpolation methods allow the calculation of unknown numerical values using known data due to spatial autocorrelation. Interpolated models tend to smooth the data, masking heterogeneity. Finally, machine learning is built on a random forest regression algorithm when the input data pass through each decision tree and the output data of all trees are averaged. Single Russian examples are represented mainly by geostatistical methods.