An aggressive subtype of breast cancer (BC) is defined as Her2/ErbB2 positive. These cancers harbor genomic amplification at the ErbB2 locus on Chromosome 17q12, this which drives potent pro-mitogenic and pro-survival signaling. Biopsies of these cancers are typically identified histopathologically by either direct demonstration of ErbB2 gene amplification, or by strong immunohistochemical staining for the ErbB2 gene product. Thus a binary readout – ErbB2 amplified (Her2+ve) or non-amplified (Her2-ve) is given. Herein we analyzed invasive breast cancer-derived RNASeq data from The Cancer Genome Atlas (TCGA) project to assess the possibility that transcriptomics can act as proxy to genome-based analysis. To our great surprise, patients clustered into three, not two separate survival groups based on differential ErbB2 transcript expression. ErbB2HI expression samples associated with worse overall survival and corresponded to the “known” Her2+ve BC subtype. ErbB2MED samples associate with better survival and but were differentially enriched for endocrine receptor positive tumors. The ErbB2LO expression cohort however associated with low overall survival and was particularly enriched for the triple-negative breast cancer (TNBC) phenotype. A systems biological analysis of differential genes expressed by the ErbB2LO cohort, demonstrated that these tumors express high levels of both the EGF Receptor (EGFR) and one of its cognate ligands, TGFA, supporting that TNBC may be driven at least in part by TGFA-EGFR signaling. A strong interferon gamma gene signature was uncovered particularly in ErbB2LO tumors, this also associating with lymphocyte activation. However, even though these and other associations were found to be strongly enriched for the ErbB2LO cohort on average, stochastic differences were detected at the individual patient level. We thus propose that measurement of ErbB2 transcript levels can provide additional diagnostic annotation in BC to that provided by the current binary ErbB2 classification system based on measurement of gene dosage or ErbB2 immunohistochemistry. Furthermore, RNASeq data can help divulge pathway activation down to the individual patient level.