The evolving factor analysis (EFA) facilitates the detection of the emergence or loss of chemical species during a chemical reaction, a chromatographic process, a spectroelectrochemical analysis, and other experiments. To this end, EFA monitors the singular values of sequences of spectral data matrices with increasing dimensions. This paper studies the growth behavior of these singular values. It investigates properties of the EFA techniques, analyzes the general shape of the singular value curves, and gives practical remarks for an improved extraction of chemical information from EFA plots.