Enzymes used during washing in laundry detergents have become a universal tool to lower energy consumption and to generate a broad, consumer-relevant, cleaning effect. However, the stability of these enzymes remains a major obstacle, particularly in liquid products, due to increased interaction between the enzymes and the other components of the detergent. The process of formulation involves extensive shelf-life stability studies where residual enzyme activities are correlated with formulation variations. As a way to improve the formulation process, we evaluated the possible use of differential scanning calorimetry (DSC) as a tool to predict enzyme stability in liquid detergents. Thus, residual enzyme activity after incubation in a multitude of formulations was determined and compared to thermodynamic data obtained by DSC. The enzymes tested were a protease, an alpha-amylase and a lipase. We found a strong linear correlation between DSC-derived data, in particular T max (temperature at peak maximum of the transition from the folded to unfolded state) and enzyme activity studies with R 2 -values: 0.98 (protease), 0.99 (amylase) and 0.98 (lipase), respectively. Thus, a higher T max for the same enzyme in a particular formulation is directly proportional to longer storage stability. These results suggest a new way of greatly accelerating this type of formulation study, allowing estimation of enzyme compatibility with a specific formulation on a daily, rather than the weekly or monthly basis used at present.