A method is developed for estimating returns to food price reporting and store selection preference changing programs. The approach is demonstrated by an example. While the extimated return to preference changing is small, the direct savings for price reporting can be substantial. Further replications are required to verify these results.