Herein, we introduce the first relative single-particle inductively coupled plasma mass spectrometry (spICP-MS) approach where size calibration is carried out using the target NP itself measured under different instrumental conditions without external dependence on the complex and prone-to-error determination of transport efficiency or mass flux calibrations, in contrast to most spICP-MS approaches. The simple approach proposed allows determining gold nanoparticle (AuNP) sizes, with errors ranging from 0.3 to 3.1% (corroborated by HR-TEM). It has been demonstrated that the changes observed in the single-particle histograms obtained for a suspension of AuNPs under different sensitivity conditions (n = 5) are directly and exclusively related to the mass (size) of the target AuNP itself. Interestingly, the relative nature of the approach shows that once the ICP-MS system has been calibrated with a generic NP standard, it is no longer necessary to repeat the calibration for the size determination of different unimetallic NPs carried out along time (at least 8 months), independently of their size (16-73 nm) and even nature (AuNP or AgNP). Additionally, neither the NP surface functionalization with biomolecules nor protein corona formation led to significant changes (relative errors slightly increased 1.3- to 1.5-fold, up to 7%) in the NP size determination, in contrast to conventional spICP-MS approaches where relative errors increased 2- to 8-fold, up to 32%. This feature could be especially valuable for the analysis of NPs in real samples without the need of matrix-matched calibration.