In GC–MS, typically, the majority of masses are of no significance. The aim of this paper is to show how the influence of increasing numbers of masses affect the information content of GC–MS data. Two closely eluting peaks arising from salbutamol and clenbuterol are analysed. Principal component analysis is performed using 10 and 50 masses. The patterns formed using two principal components are compared by procrustes analysis, involving scaling, rotating and reflecting the data. The influence of increasing the number of masses is discussed. The change in pattern is quantified using the root mean square difference between the scores using 50 masses and a smaller number of masses.