BACKGROUNDDetermination of composition and physicochemical parameters of natural products requires dedicated, often laborious and expensive, analytical protocols. Different spectroscopic techniques, in conjunction with chemometrics, seem to have a considerable potential in direct analysis of raw plant material and foods, without any chemical treatment.RESULTSFluorescence spectroscopy and three vibrational spectroscopy techniques were applied to determine total polyphenols content, antioxidant activity and macronutrient levels in red‐ and purple‐fleshed potato varieties. Excitation‐emission matrix (EEM) fluorescence, FT Raman, attenuated total reflection infrared (FTIR ATR) and near‐infrared (NIR) spectra were recorded for the freeze dried samples. Combining spectral data and the results of reference analyses, partial least squares regression (PLS) models were constructed for each parameter studied. For polyphenols and antioxidant activity, quantification errors found for validation samples amounted to 3.74–5.04% and 4.75–6.35%, respectively, whereas macronutrient analysis gave errors in the 3.45–4.55%, 3.09–5.30%, and 5.10–8.58% ranges for starch, protein, and sugars determinations, respectively.CONCLUSIONThe obtained results demonstrate that different spectroscopic techniques in combination with multivariate modeling allow simultaneous determination of various parameters of plant samples based on a single sample spectra. They can effectively replace commonly used protocols of food product analysis requiring sample dissolving and extraction of the compounds of interest.This article is protected by copyright. All rights reserved.