Gas chromatography-mass spectrometry (GC-MS) and gas chromatography-ion mobility spectrometry (GC-IMS) are two powerful separation and detection techniques for non-targeted analysis (NTA) of volatile compounds. Combined with multivariate chemometric analysis, they have been successfully applied to several areas in chemical, environmental, food, flavor and medical sciences. The recorded raw GC-MS and GC-IMS data include high numbers of variables (features) due to the high scan speeds of the instrument. Additionally, NTA approaches, by design, record more data than required. Therefore, reducing the number of features is a key step in any chemometric pipeline to reduce overfitting, overlong training times and model complexity. There are various feature engineering strategies in the literature for GC-MS and GC-IMS data which make it more difficult for users to decide which method is more appropriate for their cases. Therefore, the objective of this tutorial paper is presenting the idea of feature selection and feature extraction techniques in GC-MS and GC-IMS data and then, comparing the performance of different strategies in NTA approaches. Additionally, the effect of feature selection and extraction strategies in pattern recognition problems will be discussed with some informative examples. We will demonstrate workflows based on available tools that can be used to ensure that every researcher can follow our tutorial.