Biochemical markers of bone turnover may be useful to monitor patients taking hormone replacement therapy (HRT). The aim of this study was to assess the utility of markers in monitoring HRT by comparing the response of a large panel of markers to HRT with their within subject variability. We measured the response of markers to transdermal estradiol in 11 postmenopausal women over 24 weeks. We measured the within subject variability of markers in 11 untreated healthy postmenopausal women over the same period. The mean decrease in markers of bone formation after 24 weeks treatment ranged from 19% for procollagen type I C-terminal propeptide (PICP) to 40% for procollagen type I N-terminal propeptide (PINP). The mean decrease in markers of bone resorption ranged from 10% for tartrate-resistant acid phosphatase (TRAP) to 67% for C-terminal cross-linked telopeptide. The least significant change (LSC at p < 0.05), calculated from the within subject variability in the untreated group, was used to define response. LSC for osteocalcin was 21%, bone alkaline phosphatase 28%, PICP 24%, PINP 21%, type I collagen telopeptide 28%, TRAP 17%, urinary calcium 90%, hydroxyproline 75%, total deoxypyridinoline 47%, free pyridinoline 36%, free deoxypyridinoline 26%, N-terminal cross-linked telopeptide 70%, and C-terminal cross-linked telopeptide 132%. The greatest number of responders after 24 weeks of treatment were found using PINP and osteocalcin (9 each), and free deoxypyridinoline (8 each) and total deoxypyridinoline (7 each). Lumbar spine bone mineral density defined four patients as responders. The ability to detect a response differs between markers and is not dependent on the magnitude of response to therapy.