Plant cell walls are composed of a number of coextensive polysaccharide-rich networks (i.e., pectin, hemicellulose, protein). Polysaccharide-rich cell walls are important in a number of biological processes including fruit ripening, plant–pathogen interactions (e.g., pathogenic fungi), fermentations (e.g., winemaking), and tissue differentiation (e.g., secondary cell walls). Applying appropriate methods is necessary to assess biological roles as for example in putative plant gene functional characterization (e.g., experimental evaluation of transgenic plants). Obtaining datasets is relatively easy, using for example gas chromatography–mass spectrometry (GC-MS) methods for monosaccharide composition, Fourier transform infrared spectroscopy (FT-IR) and comprehensive microarray polymer profiling (CoMPP); however, analyzing the data requires implementing statistical tools for large-scale datasets. We have validated and implemented a range of multivariate data analysis methods on datasets from tobacco, grapevine, and wine polysaccharide studies. Here we present the workflow from processing samples to acquiring data to performing data analysis (particularly principal component analysis (PCA) and orthogonal projection to latent structure (OPLS) methods).