Response surface methodology was employed to optimize culture medium for production of lipase with Rhodotorula sp. MTCC 8737. In the first step, a Plackett-Burman design was used to evaluate the effects of different inducers qualitatively. Of all the seven inducers tested, soybean oil showed significant influence on the lipase production. Further, response surface studies were conducted to quantitatively optimize by considering linear, interactive, and quadratic effects of test variables. A novel approach was proposed to optimize the lipase production system by optimizing the responses in terms of yield kinetics rather than optimizing the direct responses like lipase titer and biomass growth. The coefficient of determination (R-2) calculated for Y-P/S (0.769), Y-P/X (0.799), and Y-X/S (0.847) indicated that the statistical model could explain 76.9%, 79.99%, and 84.7% of variability in the response.