Traditional approaches to recommendations emphasize soil testing (for inputs other than nitrogen), a univariate (e.g. multiple rates of a specific nutrient) approach to field experimentation and statistical analyses, and tabular summarizations / translations that require limited farmer data to determine a site or soil specific input amounts for a given nutrient and farm. Recognized challenges to this approach include inherently limited inference space and insensitivity to any other influencing or modifying environmental and/or management factors not considered in the initial studies. Furthermore, the lack of data standards and a tradition of publishing synthesized but not raw data make it impossible to aggregate existing with new data to broaden or improve recommendations with strategically designed ancillary studies. Given the wealth of measurement and monitoring equipment currently collecting data on or relevant to US farms and the prevalence, low cost and potential utility of data management tools and cyberinfrastructure, implementation of alternative recommendation frameworks offer significant promise to overcome these challenges and lead to more robust recommendations. Key elements that need to be explored in recommendation development include: 1) distributed data repositories of curated and preserved data conforming to standardized vocabularies, 2) tools and workflows for disparate data aggregation and that facilitate broad repository use such as user-friendly data entry interfaces and embargoing, citation and proprietary data protection mechanisms, 3) a multivariate approach to statistical modeling, 4) cyber workspace that allows for collaborations among researchers and farmers and their advisors / networks and 5) an explicit strategy for recommendation updating and improvement with database expansion. This approach will be discussed in the context of two on-going, multi-location experiments which are a 15 yr, 5 location study to improve soil potassium test interpretation and potassium recommendations for maize and soybean in the Eastern Cornbelt and a comparative study to of novel bioenergy cropping systems where characterization of system carbon and nitrogen cycling and, thus, eventual development of N recommendations are core objectives.