The adoption of Enhanced Recovery After Cesarean is increasing, but evidence supporting individual interventions having a specific benefit to Enhanced Recovery After Cesarean is lacking. A key element in Enhanced Recovery After Cesarean is early oral intake. Maternal complications are more frequent in unplanned cesarean delivery. In planned cesarean delivery, immediate full feeding enhances recovery, but the effect of unplanned cesarean delivery during labor is not known. This study aimed to evaluate immediate oral full feeding vs on-demand oral full feeding after unplanned cesarean delivery in labor on vomiting and maternal satisfaction. A randomized controlled trial was conducted in a university hospital. The first participant was enrolled on October 20, 2021, the last participant was enrolled on January 14, 2023, and follow-up was completed on January 16, 2023. Women were assessed for full eligibility on arrival at the postnatal ward after their unplanned cesarean delivery. The primary outcomes were vomiting in the first 24 hours (noninferiority hypothesis and 5% noninferiority margin) and maternal satisfaction with their feeding regimen (superiority hypothesis). The secondary outcomes were time to first feed; food and beverage quantum consumed at first feed; nausea, vomiting, and bloating at 30 minutes after first feed, at 8, 16, and 24 hours after the operation, and at hospital discharge; parenteral antiemetic and opiate analgesia use; first breastfeeding and satisfactory breastfeeding, bowel sound, and flatus; second meal; cessation of intravenous fluid; removal of a urinary catheter; urination; ambulation; vomiting during the rest of hospital stay; and serious maternal complications. Data were analyzed using the t test, Mann-Whitney U test, chi-square test, Fisher exact test, and repeated measures analysis of variance as appropriate. Overall, 501 participants were randomized into immediate or on-demand oral full feeding (sandwich and beverage). Vomiting in the first 24 hours were reported by 5 of 248 participants (2.0%) in the immediate feeding group and 3 of 249 participants (1.2%) in the on-demand feeding group (relative risk, 1.7; 95% confidence interval, 0.4-6.9 [0.48%-8.28%]; P=.50), and the maternal satisfaction scores from 0 to 10 were 8 (6-9) for the immediate feeding group and 8 (6-9) for the on-demand feeding groups (P=.97). The times from cesarean delivery to the first meal were 1.9 hours (1.4-2.7) vs 4.3 hours (2.8-5.6) (P<.001), first bowel sound 2.7 hours (1.5-7.5) vs 3.5 hours (1.8-8.7) (P=.02), and second meal 7.8 hours (6.0-9.6) vs 9.7 hours (7.2-13.0) (P<.001). These intervals were shorter with immediate feeding. The participants were more likely to agree to recommend immediate feeding to a friend (228 [91.9%] in the immediate feeding group vs 210 [84.3%] in the on-demand feeding group; relative risk, 1.09; 95% confidence interval, 1.02-1.16; P=.009). However, at first feed for food, ate "nothing at all" rates were 10.4% (26/250) in the immediate group and 3.2% (8/247) in the on-demand group, and "eaten all" rates were 37.5% (93/249) in the immediate group and 42.8% (106/250) in the on-demand group (P=.02). Other secondary outcomes were not different. Compared with on-demand oral full feeding, immediate oral full feeding after unplanned cesarean delivery in labor did not increase the maternal satisfaction score and was not noninferior on postoperation vomiting. On-demand feeding with its emphasis on patient autonomy could be preferred, but the earliest full feeding should be encouraged and provided.