Abstract Objectives To determine the pattern of feeding milestones following primary repair of long-gap esophageal atresia (EA). Method A questionnaire based upon well established feeding milestones was used. Children after long-gap EA repair, n = 40, were compared from after primary repair to healthy children from birth, n = 102. Results The age when surveyed of the EA group and controls was different: 6.2 ± 4.7 (mean ± standard deviation) years, range 1.1–20.9, versus 2.5 ± 2.4 years, range 0.0–12.1, p = 0.00. The esophageal gap length in the EA group was 5.1 ± 1.2 cm and age at repair was 5.5 ± 5.0 months. There was no statistically significant difference between the atresia group and controls for feeding milestones; Self feeding finger foods approached significance. There was, however, greater variability in the timing of milestones in the atresia group compared to controls. Feeding milestones were negatively correlated with age at primary repair: drinking with a covered sippy cup, ρ = − 0.51, p = 0.01 and self feeding finger foods, ρ = − 0.36, p = 0.04 were statistically significant. Drinking from a cup correlated with gestational age, ρ = 0.38, p = 0.04, and negatively correlated to esophageal gap length, ρ = − 0.45, p = 0.01. Conclusions Despite delayed onset of feeding, major milestones after EA repair occurred in similar pattern to normal infants. An early referral for primary repair is beneficial for earlier acquisition of milestones for infants with long-gap EA.