Arguing for better policy design in affirmative action, this paper presents an illustrative model of a feasible alternative to caste quotas. The proposed model is evidence-based, addresses multiple sources of group and individual disadvantage (caste, region, gender and rural/urban residence), as well as interaction effects and degrees of disadvantage. Such an approach allows us to demonstrate that affirmative action is not about "appeasement" but about eliminating sources of tangible disadvantage in our unequal society.