Lymphoid malignancies are grouped and characterized by their morphology, immunophenotype, and genetic aberrations to help establish a diagnosis. Use of microarray technology enables the simultaneous determination of expression levels for thousands of genes, providing an additional powerful tool for improving disease classification. In this review, recent studies of diffuse large B-cell lymphoma (DLBCL), primary mediastinal large B-cell lymphoma, mantle cell lymphoma (MCL), and follicular lymphoma are highlighted, and the impact of gene expression profiling on the molecular diagnosis of these diseases is discussed. Based on microarray-generated gene expression profiles, outcome predictors were constructed for DLBCL and MCL. Specific expression patterns of a limited number of genes at the time of diagnosis were linked to overall survival in DLBCL and MCL. Such predictors of prognosis may eventually lead to risk-adjusted treatment of lymphomas. Specific therapeutic targets may also emerge with increased insight into the molecular features of the different lymphomas, thus illustrating the usefulness of gene expression profiling not only to improve diagnosis and classification but also to generate prognostic indicators and targets for therapy.