Within the field of molecular biology, data independent acquisition (DIA) mass spectrometry techniques such as SWATH-MS (sequential window acquisition of all theoretical fragment ion spectra) have experienced rapid growth in recent years, showing significant promise in characterising proteomic variation across a range of species, sample types and biological conditions. In particular, SWATH-MS shows significant promise in biomarker research. However, studies in human blood in particular continue to confound researchers, due to the inherent difficulty of analysing this material. In this study, we explored the role of spectral library size and composition, using a range of library sizes and composition to interrogate multiple cellular and blood serum sample sets. In addition, we explored an using entrapment approach in a DIA context, using a spectral library generated from the thermophilic archaea Pyrococcus furiosus to attempt to determine a standard of Ã¢ÂÂtrueÃ¢ÂÂ identifications. For consistent manipulation of libraries, we developed TraMLMunger, a software tool to perform arbitrary manipulations on spectral libraries. We demonstrated that library size alone had no impact on the number of peptides identified in a SWATH-MS experiment, but that library composition had a significant impact due to the target-decoy statistical model used to verify peptide hits. We suggest to researchers that a one-size-fits-all approach to library choice in a SWATH-MS study is overly reductive, and recommend a range of guidelines to choose a library based on the biological material type and the fitness function of a given study.