In this study, a pulsed mode ultrasound assisted extraction was performed for the extraction of bioactive compounds from Pomegranate (Punica granatum L.) peel. Box-Behnken experimental design combined with response surface methodology, was applied for the optimization of process conditions. The independent variables investigated were ultrasound amplitude (50–80%), sonication time (5–15 min), duty cycle (0.2–0.8), and methanol concentration (30–70%) to maximize the yield of total phenolic content, flavonoid content, total condensed and hydrolysable tannin (TCT and THT respectively), and antioxidant activity (DPPH and FRAP). The experimental results were fitted to quadratic models, and statistical analysis in terms of multiple regression, and analysis of variation was carried out for each response variable. The optimum predicted values of the response variables were obtained at 70% methanol concentration, 12.8 min of sonication treatment at a duty cycle of 0.58, and 80% ultrasound amplitude. A comparative analysis at optimized condition showed better recovery of bioactive compounds in pulsed mode, compared to continuous mode of sonication. Correlation analysis of the response variables was done to find the associations among the variables. Multivariate analysis using principal component analysis, and hierarchical cluster analysis was done for better understanding of interactions among experimental treatments and responses.