Septic shock, the excessive host immune response to pathogen infection, poses a major health concern accounting globally for approximately 20% of all deaths. Current methods to establish disease severity are unacceptably slow, unspecific and insensitive, hindering timely and effective treatment and patient care. Aiming to identify easy-to-assay glyco-signatures that may identify and guide the clinical management of the worst affected patients, we applied quantitative glycomics and glycoproteomics to serum longitudinally collected from septic shock survivors (n = 29) and non-survivors (n = 8). Glycomics of all 134 serum samples (2-7 samples/patient, daily sampling until recovery or death) revealed significant N-glycome dynamics in both septic shock survivors and non-survivors. Interestingly, unsupervised clustering of the serum N-glycome upon ICU admission (day 1) indicated survivorship-specific glyco-signatures. We therefore trained a random forest model using the serum N-glycome data from all ICU samples except for day 1 and tested the model using the clinically informative day 1 data. Excitingly, the model accurately classified survivorship outcomes of 35 of 37 patients (specificity 94.6%) and correctly predicted 6 of 8 non-survivors (sensitivity 75%). Targeted N-glycome analysis revealed raised levels of Lewis fucosylation in non-survivors relative to survivors already upon ICU admission. Elevated Lewis fucosylation was recapitulated by comparative glycoproteomics that also identified α-1-acid-glycoprotein as the principal carrier of Lewis fucosylation. Enabled by integrated -omics approaches, this study has laid a foundation for improved clinical management of septic shock patients by uncovering easy-to-assay glyco-signatures that early in the disease course accurately identify individuals with poor survival outcomes.