In the absence of an intervening antiresorptive agent, cyclic administration of teriparatide does not increase bone mineral density (BMD) more than standard daily therapy. Because denosumab is a potent antiresorptive agent with a rapid off‐effect, we hypothesized that it might be the optimal agent to help maximize bone gains with cyclic teriparatide. In this 3‐year protocol, 70 postmenopausal women with osteoporosis were randomized to 18 months of teriparatide followed by 18 months of denosumab (standard) or three separate 12‐month cycles of 6 months of teriparatide followed by 6 months of denosumab (cyclic). BMD (dual‐energy X‐ray absorptiometry [DXA]) measurements of lumbar spine (LS), total hip (TH), femoral neck (FN), and 1/3 radius (RAD) were performed every 6 months and total body bone mineral (TBBM) at 18 and 36 months. Baseline descriptive characteristics did not differ between groups except for a minimal difference in LS BMD but not T ‐score (mean age 65 years, mean LS T ‐score − 2.7). In the standard group, BMD increments at 36 months were:​ LS 16%, TH 4%, FN 3%, and TBBM 4.8% (all p  < 0.001 versus baseline). In the cyclic group, 36‐month BMD increments were similar:​ LS 12%, TH 4%, FN 4%, and TBBM 4.1% (all p  < 0.001 versus baseline). At 36 months, the LS BMD increase with standard was slightly larger than with cyclic (p = 0.04), but at 18 months, in the cyclic group, there was no decline in RAD or TBBM (p = 0.007 and < 0.001, respectively, versus standard). Although the cyclic regimen did not improve BMD compared with standard at 36 months, there appeared to be a benefit at 18 months, especially in the highly cortical skeletal sites. This could be clinically relevant in patients at high imminent risk of fracture, particularly at nonvertebral sites.