Despite the raising preoccupation, the critical question of how the plant community is composed belowground still remains unresolved, particularly for the conservation priority types of vegetation. The usefulness of metabarcoding analysis of the belowground parts of the plant community is subjected to a considerable bias, that often impedes detection of all species in a sample due to insufficient DNA quality or quantity. In the presented study we have attempted to find environmental factors that determine the amount and quality of DNA extracted from total plant tissue from above- and belowground samples (1,000 and 10,000 cm2). We analyzed the influence of land use intensity, soil properties, species composition, and season on DNA extraction. The most important factors for DNA quality were vegetation type, soil conductometry (EC), and soil pH for the belowground samples. The species that significantly decreased the DNA quality were Calamagrostis epigejos, Coronilla varia, and Holcus lanatus. For the aboveground part of the vegetation, the season, management intensity, and certain species — with the most prominent being Centaurea rhenana and Cirsium canum — have the highest influence. Additionally, we found that sample size, soil granulation, MgO, organic C, K2O, and total soil N content are important for DNA extraction effectiveness. Both low EC and pH reduce significantly the yield and quality of DNA. Identifying the potential inhibitors of DNA isolation and predicting difficulties of sampling the vegetation plots for metabarcoding analysis will help to optimize the universal, low-cost multi-stage DNA extraction procedure in molecular ecology studies.