Clinical trials have shown that a variety of aromatase inhibitor strategies modestly reduce the risk of breast cancer recurrence over 5 years of tamoxifen for postmenopausal women with endocrine-sensitive disease. This review will discuss the use of predictive biomarkers to tailor postmenopausal endocrine therapy.In spite of early promise, recent studies suggest that quantitative evaluation of estrogen receptor, progesterone receptor, and human epidermal growth factor receptor-2 expression do not identify a subgroup of patients with differential benefit from an aromatase inhibitor over tamoxifen therapy. Estrogen receptor-positive tumors with increased proliferation, evaluated by Ki-67, may be selectively responsive to aromatase inhibitor therapy. Growing evidence suggests that inherited polymorphisms in tamoxifen-metabolizing enzymes identify a subgroup of patients with poor outcome under tamoxifen therapy that might best be treated with an aromatase inhibitor. The emergence of early treatment-related side effects on either aromatase inhibitor or tamoxifen, such as hot flashes and anthralgias, has been linked to a decreased risk of recurrence, although additional validation is needed.There are no robust clinically available tools that can be used to reliably identify patients that are likely to be selectively responsive to an aromatase inhibitor over tamoxifen alone. The decision to use an aromatase inhibitor in clinical practice should be primarily guided by consideration of recurrence risk and contraindications to tamoxifen.