Abstract: Nitrogen use efficiency (NUE) is a key indicator for evaluating the crop uptake of nitrogen (N) fertilizer applied to cropland. The principal approaches for quantifying NUE include N difference and ¹⁵N tracer. However, these approaches significantly underestimated the actual efficiency of N fertilizer, as these approaches failed to consider the residual and legacy effects of N fertilizer in soil. Therefore, researchers all over the world have been investigating the more actual NUE. The NUE calculated by improvement methods is analogous to the actual NUE, but higher than the N difference or the ¹⁵N tracer approaches. This can be attributed to the differing forms in which various improvement methods incorporate the legacy effects of N fertilizer, thereby compensating for the underestimation of NUE using the N difference or the ¹⁵N tracer approaches in short-term experiments. In this review, the actual NUE is defined as the percentage of N fertilizer uptake by the crop aboveground to the amount of N fertilizer inputs, while completely consuming the residual N fertilizer in soil. The accumulative N recovery efficiency is 61.3%−65.3% of a 28-year upland cropping system in France, and 38.6%−43.0% of a 17-year paddy-upland rotation system in the Taihu Lake region of China. We obtained the data from the two long-term field experiments using the Digitizer tool of Origin 2021, including the crop ¹⁵N uptake and the residual ¹⁵N in soil. The NUE was calculated based on the N residual in the topsoil (0−20 cm), 1 m, and 2 m soil, respectively. We analyzed and evaluated the rationality of the computed results of the soil N balance and N fertilizer efficiency approaches by comparing them to the actual NUE. The soil N balance and N fertilizer efficiency approaches significantly overestimated NUE when calculated by two meters of residual N in the soil. The surface (0−20 cm) soil residual N was found as a calculation parameter for estimating the actual NUE in these two approaches (61%−74%). However, the N fertilizer efficiency (41%−52%) was the only approach that accurately estimated the actual NUE in the Taihu Lake region of China. Therefore, the comparison of NUE across plots or crop systems should be based on a consistently chosen quantification approach to avoid inherent biases introduced by different approaches. In general, the N fertilizer efficiency approach (both crop absorption and soil residual nitrogen are considered to be utilized) achieves near the actual NUE in short-term experiments, demonstrating extensive application value.