ABSTRACT Modern breeding programs aim to increase the effective use of nitrogen (N) in cereal grain crops such as wheat ( Triticum aestivum L.) to reduce the environmental footprint of agriculture. The effective use of nitrogen may be assessed by measuring N utilization efficiency (NUtE, grain yield to N uptake ratio, both determined at harvest) or by measuring N nutrition index at anthesis (NNI, actual to critical N concentration ratio). Our main objective was to compare NUtE and NNI as metrics for phenotyping effective use of N in wheat. We used data from 11 field experiments evaluating wheat cultivars in contrasting N levels, including a revisionist analysis with historical cultivars. We found that NUtE had nearly no relationship with yield and was negatively related to grain protein concentration. In contrast, we found that NNI was positively related to yield and grain number, and was slightly related to grain protein concentration. Regarding the historical analysis, we found that the efficient use of N increased, but NUtE‐based analyses seemed to estimate greater improvements that NNI‐based analyses. In summary, NUtE may provide overoptimistic results on phenotyping and should be replaced by allometric‐based, plant physiology‐derived metrics such as NNI.