Geranium wilfordii Maxim. has long been employed in traditional East Asian medicine owing to its anti-inflammatory 

and antioxidant properties. Despite its therapeutic potential, previous studies have largely focused on biological activity 

evaluation, while systematic studies addressing extraction process optimization and standardization for reproducible and 

industrially applicable use remain scarce. In this study, response surface methodology (RSM) with a Box–Behnken design 

was applied to establish a process-oriented optimization framework for the extraction of representative ellagitannins from 

G. wilfordii. The effects of extraction temperature, ethanol concentration, and extraction time on extraction yield, corilagin

content, and geraniin content were evaluated. Regression models were constructed from the experimental data, and 

multi-response optimization was performed using the desirability function. Validation experiments were conducted under 

the predicted optimal conditions. Extraction temperature and ethanol concentration significantly affected extraction yield 

and corilagin content, whereas geraniin content was strongly dependent on ethanol concentration. The regression models 

showed acceptable adequacy with R² values of 82.96–90.64% and adjusted R² values of 67.62–82.22%. The predicted 

optimal conditions were 30 °C, 60% ethanol, and 14.67 h, yielding predicted values of 16.46% yield, 11.80 mg/g corilagin,

and 60.9 mg/g geraniin. Validation experiments yielded 13.11% yield, 9.97 mg/g corilagin, and 44.05 mg/g geraniin, 

demonstrating reasonable agreement with the predicted values. These results demonstrate that RSM provides a robust and 

reproducible framework for extraction process optimization, offering a scientific basis for the standardization and industrial

utilization of G. wilfordii extracts as functional ingredients.