A CD44(-)/CD24(+) phenotype is a poor prognostic marker in early invasive breast cancer. Breast cancer cells with high CD44 and low or absent CD24 (i.e. CD44(+)CD24(-)/low phenotype) are reported to have stem cell features. However, the clinical impact of CD24 and CD44 expression in tumours remains unclear. To explore the immunohistochemical expression of CD44 and CD24 (individually and combined) and their clinical value as prognostic and predictive markers. Immunohistochemical expression of CD24 and CD44 was studied in a large series of early primary invasive breast cancer tumours (n = 1036) prepared as a tissue microarray. Associations between the expression of each marker individually and in combination and clinico-pathological, molecular variables and patients' outcome were investigated. CD24 cytoplasmic expression was significantly associated with poor prognostic variables including high tumour grade, ER-, PR-, HER2(+), p53+ and triple negative (TN) phenotype; P < 0.05. However, CD24 expression was not significantly associated with patients' outcome. Conversely, CD44 expression was associated with favourable prognostic criteria including lower Nottingham prognostic index, ER+, HER2- and luminal phenotype; P < 0.05. Moreover, CD44 expression was found to be an independent predictor of good prognosis. In combination, the CD44(+)/CD24(-) phenotype was associated with the most favourable outcome (84 and 80% 10 year breast cancer survival [BCSS] and metastasis free survival [MFS], respectively). Contrasting this, the CD44(-)/CD24(+) phenotype was associated with the most dismal outcome (62 and 60% 10 years BCSS and MFS, respectively). CD24 and CD44 expression can individually yield prognostic data in breast cancer, but importantly, when both markers are considered; the CD44(+)/CD24(-) phenotype had the best prognosis, while the CD44(-)/CD24(+) phenotype had the worst prognosis. This shows that the relationship between basic cell biology and clinical behaviour is not always straightforward and warrants further investigations of the true clinical impact of breast cancer stem cells.