Diffuse large B cell lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma worldwide, representing approximately 30-40% of all cases in different geographic regions. Patients most often present with a rapidly growing tumour mass in single or multiple, nodal or extranodal sites. The most common type of DLBCL, designated as not otherwise specified, represents 80-85% of all cases and is the focus of this review. There are also rare types of lymphoma composed of large B-cells, in aggregate about 15-20% of all neoplasms that are sufficiently distinctive to recognise separately. DLBCL not otherwise specified (referred to henceforth as DLBCL) is a heterogeneous entity in terms of clinical presentation, genetic findings, response to therapy, and prognosis. A major advance was the application of gene expression profiling (GEP) to the study of DLBCL which further clarified this heterogeneity and provided a rationale for subdividing cases into groups. The most popular system divides cases of DLBCL according to cell-of-origin into germinal centre B-cell like (GCB) and activated B-cell like (ABC) subtypes, with about 10-15% of cases being unclassifiable. Patients with the GCB subtype usually have better prognosis than patients with the ABC subtype. Although cell-of-origin is useful for predicting outcome, the GCB and ABC subtypes remain heterogeneous, with better and worse prognostic subsets within each group. Next generation sequencing (NGS) analysis of DLBCL has facilitated global identification of numerous and diverse genetic abnormalities in these neoplasms and has shown that GCB and ABC tumours have different mutation profiles. Although the therapy of patients with DLBCL is an active area of research, the current 5-year overall survival rate is 60-70% using standard-of-care frontline therapy. A precision medicine approach for the design of new therapies based on molecular findings in DLBCL is likely the best path forward. As pathologists, our role has expanded beyond diagnosis. We must perform a complete work-up of DLBCL cases. In addition to our traditional role in establishing the diagnosis, we need to analyse markers that provide information regarding prognosis and potential therapeutic targets. We also must ensure that adequate tissue is triaged for molecular studies which are essential for designing therapy regimens, particularly in the setting of disease relapse.