Saponins are triterpenic or steroidal glycosides present in many plants, as juá (Ziziphus joazeiro) and sisal (Agave sisalana), representatives of the Brazilian agrobiodiversity. Due to their properties, saponins are finding application in fine chemistry and pharmaceutical industry. Thus, the study of the extraction of these components from their natural matrixes and their optimization is of interest. The objective of this work is to evaluate the performance of the extraction of saponins from juá and sisal, using cholinium salts and deep eutectic solvents (DES) as adjuvants to aqueous and hydroalcoholic solutions. The goal is to simultaneously optimize the saponin extraction efficiency and the selectivity between saponins and phenolic compounds from these two plant matrixes. For this purpose, 13 cholinium-based ionic liquids and salts and 9 DES were used as adjuvants. It was observed that DES performed better in the proposed extractions and the optimal extraction mixture composition was determined using a mixture-design statistic method. Using this information, other experimental parameters, such as extraction time, raw material/solvent ratio and temperature were also evaluated and optimized. In the end, we show that it is possible to increase the extraction efficiency and selectivity of saponins from juá in 170 % and 18.39 and for sisal 245 % and 10.34, respectively, by the addition of a selected DES.