Researchers working on a USDA-sponsored research project are currently conducting approximately one hundred large-scale, on-farm agronomic field trials in seven countries and seven US states.  Each experiment randomizes input application rates on full fields no less than 35 hectares in area.  The methodology of their experiments is to use precision technology to design and conduct the trials; farmers can implement the trials with very little bother.

We report the results of research developing an automatized analytical engine to be used to analyze the data of thousands of such on-farm field trials per year.  Research results will be of significant importance to the data-intensive farm management industry, particularly for the management of nitrogen (N) fertilization rates.  This engine (written in R code) will take the high-quality, high-volume agronomic field trial data and automatize the (1) cleaning of the raw data, (2) statistical analysis of those data to estimate technical aspects of yield response to factors of production, (3) economic analysis that provides to farmers profitable input management advice, and (4) efficient conveyance of the analytical results to crop consultants, who then use it to provide their farmer-clients with practicable input management advice.  If successful, this engine will greatly increase the actual “data-intensity” of the data-intensive farm management industry, and thereby increase farmers’ N fertilizer application efficiency.  The innovation is original in that it actually provides “data-intensive farm management” advice; the majority of products currently on the market claim to do so, but are always based inadequate data, and often on discredited, yield-based algorithms that were developed in the 1960s to provide farmers with data-extensive farm management advice.  The innovation is novel in that it will use data generated by randomized experimentation on farm fields to provide profitable information about yield response on those same fields; not small-plot data on a research farm, and not sparse split-plot data, but rather dense, "checkerboard" trial data that permit identification of differing crop yield response to inputs within the field, and thus providing the information to use variable rate input application technology efficiently. The analytical engine is unusual because it can be used to prove that its use has increased a farmer’s profits; current market strategies depend on anecdotal evidence from farmers (paid to appear in advertisements), and what we call “bells and whistles” software—products actually based on very little data, but that provide beautiful graphics and verbiage about "Big Data.”  The proposed innovation will disrupt the current market's domination by “bells and whistles.”