Historically, the agronomic focus of crop mineral nutrition has yielded responses to individual elements (N, P, K...) to determine the economically optimum fertilization rates. This prognostic approach required several parameters for crops, climates, and soils that are often estimated with large uncertainty leading to over-fertilization and environmental problems in some systems (e.g., maize in China), and under-fertilization and soil mining in other systems (e.g., wheat in Australia). In this review, an alternative approach is developed for reducing the uncertainty intrinsically linked to this prognostic approach. Our approach is based on four propositions: (1) the evidence of an allometry between the metabolic shoot mass (scaling with leaf area) and the structural shoot mass (supporting and vascular tissues) within plants that allows the formulation of critical N dilution curves and the determination of the Nitrogen Nutrition Index (NNI) for estimating the N nutrition status of field crops; (2) the co-regulation of crop N uptake dynamics by both soil N supply and crop N demand in relation with its growth capacity that allows a better, more generalizable estimation of timing and rate of fertilizer; (3) a better understanding of the effects of genotype-environment-management interactions on N use efficiency in cropping systems reducing then drastically uncertainties linked to the classical prognostic approach for N fertilization; (4) as P and K also relate allometrically with biomass, P and K concentrations can be directly related to N concentration for the formulation of a multi-element diagnosis of crop nutrition. Here, we develop the theoretical background supporting these four propositions and outline implications for both fertilization management and crop phenotyping.