The direct conversion of dinitrogen to nitrate is a dream reaction to combine the Haber–Bosch and Ostwald processes as well as steam reforming using electrochemistry in a single process. Regrettably, the corresponding nitrogen oxidation (NOR) reaction is hampered by a selectivity problem, since the oxygen evolution reaction (OER) is both thermodynamically and kinetically favored in the same potential range. This opens the search for the identification of active and selective NOR catalysts to enable nitrate production under anodic reaction conditions. While theoretical considerations using the computational hydrogen electrode approach have helped in identifying potential material motifs for electrocatalytic reactions over the last decades, the inherent complexity of the NOR, which consists of ten proton-coupled electron transfer steps and thus at least nine intermediate states, poses a challenge for electronic structure theory calculations in the realm of materials screening. To this end, we present a different strategy to capture the competing NOR and OER at the atomic scale. Using a data-driven method, we provide a framework to derive generalized design criteria for materials with selectivity toward NOR. This leads to a significant reduction of the computational costs, since only two free-energy changes need to be evaluated to draw a first conclusion on NOR selectivity.