The aim of this study was to test Ovsynch (OVS) versus modified OVS (decreasing the interval between first GnRH and PGF(2 alpha) to 6 d) protocols on pregnancy per artificial insemination (AI) and OVS outcomes in cyclic dairy cows. Cyclic cows (n = 920) were assigned to 1 of 2 groups: the OVS7 group (n = 459) received the OVS protocol [GnRH treatment, PGF(2 alpha) treatment 7 d later, a second GnRH (GnRH2) treatment 56 h later, and timed AI (TAI) 16 to 18 h after the GnRH2 treatment], and the OVS6 group (n = 461) received a modified OVS protocol, in which the interval between the first GnRH and PGF(2 alpha) was decreased to 6 d (GnRH treatment, PGF(2 alpha) treatment 6 d later, GnRH2 treatment 56 h later, and TAI 16 to 18 h after the GnRH2 treatment). The response to the first GnRH of OVS was similar between OVS7 (54.5%, 250/459) and OVS6 (54.2%, 250/461) groups. The ovulatory response to GnRH2 of OVS was higher in OVS6 (91.3%, 421/461) than OVS7 (84.5%, 388/459). The follicle size (mean +/- standard error of the mean) at the time of TAI was smaller in OVS6 (15.23 +/- 0.11 mm) than OVS7 (16.04 +/- 0.11 mm). When all cows were evaluaed, the pregnancy per AI at 31 d tended to be lower in OVS6 (38.0%, 175/461) than in OVS7 (43.8%, 201/459). Moreover, the pregnancy per AI at 31 d was lower in OVS6 (40.9%, 172/421) compared with OVS7 (50.3%, 195/388) in synchronized cows. In conclusion, although the modified OVS protocol decreased the follicle size at the time of AI and increased the ovulatory response to GnRH2 of OVS, it unexpectedly reduced fertility in cyclic lactating dairy cows.