Protein glycosylation is an important but poorly understood aspect of bacterial physiology. Over the last decade significant strides have been made in the characterisation of bacterial glycosylation systems with mass spectrometry (MS) emerging as an indispensable tool for microbial glycoproteomics. While early studies focused on confirming the presence of glycosylation events little attention has been given to characterising the impacts of these systems or mapping the specific sites of glycosylation within proteins. Focusing on the oligosaccharyltransferases of the protein glycosylation loci Ligase (pglL) O-linked glycosylation systems, we have begun to explore the preferences of these enzymes as well as how these modifications impact microbial physiology using MS-based systems biology. Using members of the Burkholderia genus as a model our work has shown the glycoproteomes of these species are far larger than once thought, with >100 proteins targeted for O-linked glycosylation. While a significant proportion of the glycoproteome has no known functions quantitative proteomics has revealed multiple glycoproteins appear to require glycosylation for stability and are extremely sensitive to glycosylation inhibition or silencing. Surprisingly precise glycosylation site mapping has also revealed an unexpected preference in the sites subjected to glycosylation across bacterial species with PglL glycosylation occurring nearly exclusively at serine residues. While in vivo studies have demonstrated glycosylation can occur at Threonine residues glycosylation efficiency is extremely poor even with overexpression of PglLs. This preference for glycosylation at Serine residues is observed across a range of bacterial glycoproteomes including the pathogens Neisseria gonorrhoeae and Acinetobacter baumannii. Excitingly we also find different PglL enzymes have different targeting repertoires supporting this family of enzymes are serine preferring oligosaccharyltransferases yet glycosylation compatibility is more nuanced then initially suspected. Combined these studies highlight how glycoproteomics can improve our understanding of bacterial glycosylation revealing properties overlooked even after nearly 20 years of study.