The levels of goat milk and soy milk on the basis of cow milk were optimized on the basis of sensory evaluation using Central Composite Rotatable Design (CCRD) of Response Surface Methodology (RSM). Total 13 experiments were carried out as per the design matrix suggested by RSM in standard order. All analysis were performed with Design Expert Software (8.0.4.1) and significance was defined at p<0.01. Optimization was done using optimization command. The goals were set for obtaining best combinations. Only one solution was obtained, which was in accordance with the objective set for the study. The predicted values for sensory response of the optimized formulations were 8.66, 8.51, 8.35, 8.01 and 8.41 for flavour, body and texture, mouthfeel, colour and appearance and overall acceptability, respectively using 9 point Hedonic scale. Triplicate samples were prepared using the optimized formulation and were evaluated for all the sensory attributes. The predicted scores were compared with actual scores. The calculated 't' value for all parameters were less than table values indicating that there was no significant difference (p>0.05) between the predicted and actual values of sensory scores. It is therefore confirmed that the optimized combination of goat milk, soy milk and cow milk was most suitable for preparation of rosogolla in terms of sensory characteristics of the product.