This paper employs a Bayesian hierarchical approach to estimate individual expected performance of market advisory programs in corn and soybeans. This estimation procedure is a conservative approach compared to traditional estimation, since it reduces estimation error in the expected gains from following top-performing advisory programs. Three versions of the model are estimated. The first combines information across the entire sample, while the second includes skeptical beliefs based on the efficient market hypothesis. The third divides programs into two groups based on the degree of activeness in marketing recommendations. Results indicate that even when skeptical beliefs are incorporated into the model a few programs in corn and several programs in soybeans appear to be better marketing alternative compared to a naïve strategy that mimics the market benchmark. More specifically, a skeptical farmer can expect to increase the price received for corn by 1% and the price received for soybeans by 5% following the single top-ranked program.