We revisit the main steps to derive the Ziman resistivity formula (ZRF), a frequently used approach in previous literature to perform first-principles calculations on the electron−phonon (e-ph) scattering limited resistivity of realistic materials. We find that two kinds of crucial approximations are required to reach the final form of ZRF. One is the neglect of the energy dependence of the density of states (DOS) and the average of the squared electron velocity in the Fermi shell, a narrow energy range around the Fermi level. The other one is the replacement of the weight factor for large-angle scattering in ZRF with that used in the conventional expression of momentum relaxation time. By setting up a toy model of an e-ph coupling system with a band edge lying in the Fermi shell, we find that the first approximation does not damage the validity of ZRF seriously. However, the weight factor of large-angle scattering used in the current form of ZRF makes the calculated resistivity inaccurate. Instead of the current form, we suggest that the original form of the weight factor of large-angle scattering during the derivation of ZRF should be retained. Our conclusion is further demonstrated by selecting plumbene to perform a case study on the level of firstprinciples calculations of intrinsic resistivity. We suggest a modified ZRF which can give the numerical result of the intrinsic resistivity of realistic materials with high precision even when the Fermi level meets a band edge.