Improved biomarkers are needed for pediatric inflammatory bowel disease. Here we identify a diagnostic lipidomic signature for pediatric inflammatory bowel disease by analyzing blood samples from a discovery cohort of incident treatment-na & iuml;ve pediatric patients and validating findings in an independent inception cohort. The lipidomic signature comprising of only lactosyl ceramide (d18:1/16:0) and phosphatidylcholine (18:0p/22:6) improves the diagnostic prediction compared with high-sensitivity C-reactive protein. Adding high-sensitivity C-reactive protein to the signature does not improve its performance. In patients providing a stool sample, the diagnostic performance of the lipidomic signature and fecal calprotectin, a marker of gastrointestinal inflammation, does not substantially differ. Upon investigation in a third pediatric cohort, the findings of increased lactosyl ceramide (d18:1/16:0) and decreased phosphatidylcholine (18:0p/22:6) absolute concentrations are confirmed. Translation of the lipidomic signature into a scalable diagnostic blood test for pediatric inflammatory bowel disease has the potential to support clinical decision making. Diagnostic blood-based biomarkers of pediatric IBD are limited. Here, the authors demonstrate a diagnostic lipidomic signature, comprising only of two molecular lipids. Translation of this signature into a scalable test has the potential to support clinical decision making.