Lactating Holstein cows (n = 711) on a commercial dairy farm in Wisconsin received a hormonal synchronization protocol to initiate first timed artificial insemination (TAI) on the following postpartum schedule: two injections of 25 mg PGF2alpha at 32 +/- 3 d and 46 +/- 3 d (Presynch); 100 microg GnRH at 60 +/- 3 d; 25 mg PGF2alpha at 67 +/- 3 d; and 100 microg GnRH + TAI at 69 +/- 3 d (Ovsynch). At first TAI, cows were randomly assigned to initiate the first GnRH injection of a hormonal protocol for resynchronization of ovulation (Resynch; 100 microg GnRH, d 0, 25 mg PGF2alpha, d 7, 100 microg GnRH + TAI, d 9) at 19 (D19), 26 (D26), or 33 d (D33) after first TAI to set up a second TAI service for cows failing to conceive to Ovsynch. Overall pregnancy rate per artificial insemination (PR/AI) to Ovsynch assessed 68 d after TAI was 31% and did not differ among treatment groups. For Resynch, PR/AI was assessed 26 d after TAI for D19 and D26 cows and 33 d after TAI for D33 cows. Overall PR/AI to Resynch was 32%. However, the PR/AI for D26 (34%) and D33 (38%) cows to Resynch was greater than for D19 cows (23%). Cows with a CL at the PGF2alpha injection (D19 cows) or at the first GnRH injection (D26 + D33 cows) of Resynch exhibited greater PR/AI to Resynch compared with cows without a CL. Survival analysis (failure time) of cows in the D26 and D33 treatment groups across the first three TAI services did not differ statistically. Although administration of GnRH to pregnant cows 19 d after first TAI service did not appear to induce iatrogenic embryonic loss, initiation of Resynch 19 d after first TAI service resulted in a lower PR/AI compared with initiation of Resynch 26 or 33 d after first TAI service.