Even the simplest TOF-SIMS spectrum can be complex. A typical spectrum may contain hundreds of peaks, the intensities of which can be correlated. This complexity has led to the exploration and application of multivariate analysis (MVA) methods to TOF-SIMS data in an attempt to simplify analysis and maximize the information content extracted from the data. MVA of TOF-SIMS spectra has shown promise and success for spectra and images across many systems including self-assembled monolayers, proteins, and polymers. Due to this success, multivariate analysis of TOF-SIMS data is quickly becoming more standard within the surface analysis community, yet there have been no standards set regarding how to properly apply multivariate methods to TOF-SIMS data. It is important to make sure that as progress continues in the application of MVA methods to TOF-SIMS data, standards be established in the application of theses methods, and that effort be made in the optimization of these methods, and the education of the community. This talk will highlight important considerations in the use of PCA with TOF-SIMS data. Experimental design, peak selection, data preprocessing and data interpretation will be discussed as it pertains to PCA of TOF-SIMS spectra.