In standard addition calibration, the calibration range is accomplished directly in the sample to be dosed – that is to say, the calibration is carried out by successive analyses of the same sample: firstly without addition, then after an initial addition of a known quantity of the compound to be dosed, and so on until several successive additions are made so as to be able to plot the calibration line. The increasing sample volume must of course be taken into account as the additions are made in order to calculate the actual concentrations in the sample after the additions.

If the starting sample contains the compound to be dosed, the calibration line will not pass through the y-intercept. The calibration line must be extrapolated, to intersect with the x-axis for the dosed compound: this makes it possible to estimate the initial concentration (C0) of the compound in the sample.

This calibration is tedious when there are several samples to analyze, because the successive additions must be made for each of the samples; the use of the standard addition method is thus limited to samples where a matrix effect is observed.