Since the early development of Fluorescence Correlation Spectroscopy (FCS), it has been recognized that background intensity can lead to artifacts in the amplitude of the autocorrelation function (ACF) and, consequently, to inaccurate estimates of particle numbers. Here, we present a protocol for quantitative background evaluation and amplitude correction in FCS experiments, applicable to different sources of background such as detector noise, autofluorescence, and light scattering. We demonstrate the performance of our approach through three representative case studies: (i) FCS measurements of a bright fluorophore at low concentration, (ii) FCS of dim nanoparticles affected by solvent Raman scattering, and (iii) FCS performed using a confocal setup equipped with a SPAD array, where background originates from detector hot pixels. These examples represent typical experimental conditions in which background signals compromise quantitative interpretation, illustrating how our protocol restores accuracy and reproducibility in FCS analysis. By systematically identifying and correcting these effects, the proposed protocol addresses a long-standing limitation of FCS and provides a robust framework for improving the accuracy and reproducibility of quantitative fluorescence measurements.