In 2018, Shipp and colleagues from Harvard proposed a new genetic classification system for diffuse large B cell lymphoma (DLBCL)(1). At the same time, the Staudt laboratory at the National Cancer Institute (NCI) also proposed a genetic classification(2), refined in a subsequent publication(3), with subgroups that could be clearly matched to those of the Harvard study. More recently, a UK Haematological Malignancy Research Network (HMRN) sequencing study of almost a thousand patients independently reached strikingly similar conclusions over the molecular subclassification of DLBCL(4). These studies, and the degree of consensus between them, are important for those of us who treat DLBCL. They provide a potential handle with which to rationalise the biological heterogeneity that has so far blighted our ability to deploy biologically targeted agents successfully in the first-line treatment of DLBCL. From a biological perspective these studies redefine entirely how we view the classification and pathogenesis of DLBCL. However, from the perspective of a clinician faced with an individual patient with DLBCL the implications are much less clear. Indeed, significant challenges must be overcome before these complex classifications systems can be incorporated into standard clinical practice or prospective therapeutic trials.