Adsorption of extracellular enzymes to soil minerals is assumed to protect them against degradation, while modifying their activities at the same time. However, the persistence of the activity of adsorbed enzymes remains poorly understood. Therefore, we studied the persistence of cellulase and α-amylase activities after adsorption to soil amended with various amounts (+1, +5, and +10 wt.%) of three typical soil minerals, montmorillonite, kaolinite, and goethite. Soil without mineral addition (pure soil), pure minerals, and pure dissolved enzymes were used as references. Soil mineral–enzyme complexes were prepared and then incubated for 100 days; temporal changes in enzyme activities were analyzed after 0, 0.1, 1, 10, and 100 days. The specific enzyme activities (activities normalized to protein content) and their persistence (activities relative to activities at day 0) were compared to enzyme activities in solution and after sorption to the control soil. Amylase adsorption to pure minerals increased in the following order: montmorillonite > kaolinite > goethite. That of cellulase increased in the following order: goethite > montmorillonite > kaolinite. Adsorption of enzymes to soils did not increase in the same order of magnitude as the addition of reactive binding sites. Based on inverse relationships between the amount of enzyme adsorbed and the specific enzyme activity and their persistency, we showed that a limited availability of sorption sites is important for high specific activity and persistence of the enzymes. This is probably the consequence of less and weaker bonds, as compared to a high availability of sorption sites, resulting in a smaller impact on the active sites of the enzyme. Hence, we suppose that the soil mineral phase supports microorganisms in less-sorptive environments by saving energy on enzyme production, since small enzyme release could already result in sufficient activities to degrade respective target carbon substrates.