The objective of this study was to compare the effect of hormonal treatment duration through the use of intravaginal sponges impregnated with 60 mg MAP for 6 (G6; n = 10), 9 (G9; n = 10) or 12 (G12; n = 10) days. All nulliparous (n = 15) and lactating (n = 15) Saanen goats received 37.5 µg d-cloprostenol and 200 IU eCG 24 hours before sponge removal during the nonbreeding season. After sponge removal, estrus was monitored twice a day (06h00 and 18h00) by the use of fertile bucks. Transrectal ovarian ultrasonography was performed after device removal until ovulation detection. All goats were bred by artificial insemination (AI) or natural breeding (NB). All females received 250 IU hCG 5 d after AI or NB. No difference was observed among groups G6, G9 and G12 for: estrous response (90.0, 100.0 and 100.0%, respectively); interval from sponge removal to estrus (26.7±5.6, 25.2±2.8 and 25.2±3.3 hours), duration of estrus (28.5±7.1; 21.6±3.0 and 25.2±6.1 hours); ovulation rate (80.0, 100.0 and 70.0%); number of ovulations (1.9±0.2, 1.5±0.2 and 1.4±0.2); and pregnancy rate (60.0, 80.0 and 50.0%). Interval from sponge removal to ovulation was shorter in animals from G6 (44.6±2.2 hours) than for animals from G9 (52.8±2.7 hours) and G12 (60.0±4.5 hours). All treatments were effective in the induction of synchronized estrus in Saanen goats during the nonbreeding season, but ovulation occurred earlier when goats received sponges for 6 days.