Introduction Fetal growth restriction (FGR) is still a major cause of perinatal mortality despite advances in fetal ultrasound scanning. The traditional 10th birth weight percentile as a cut-off for defining FGR could miss many cases of relative FGR falsely labelled as appropriate for gestational age (AGA) by having a birth weight > 10th percentile, but actually have a reduced growth velocity possibly due to a higher growth potential of a milder degree of placental dysfunction. Objectives Our objective was to study the role of reduced growth velocity in AGA neonates and its influence on adverse perinatal outcome. Methods We performed a retrospective study of singleton AGA (birth weight > 10th centile) neonates born in the Maastricht University Medical Centre between 2011-2015, of whom fetal biometry scans were available in two periods: 18–22 weeks and 30–34 weeks of gestational age. Growth velocities of biparietal diameter (BPDv), head circumference (HCv), abdominal circumference (ACv) and femur length (FLv), were calculated in mm/week and compared between categories of birth weight percentiles (BWp) using logistic regression, one-way ANOVA and posthoc test, and correlated to a composite adverse perinatal outcome using chi-square test. Macroscopic and histological placental analysis were collected from the pathology register, and correlated to ACv and birthweight using chi-square test. Results We identified 407 AGA neonates with available neonatal outcomes: BWp 10–16 (N = 43), BWp 16–20 (N = 21), BWp 20–50 (N = 143), BWp 50–80 (N = 119), BWp > 80 (N = 81). There was a significant linear correlation between growth velocities and BWp: BPDv (B = 6651 p = 0.016), ACv (B = 4286 p = 0.0001) and FLv (B = 9657 p = 0.0001). There was a significant difference in AC velocity (in mm/week, mean ± SD) between the BWp 50-80 (11,34 ± 0,14, reference group) and BWp 10–16 (10,58 ± 0,14 p = 0.002), BWp 16–20 (10,45 ± 0,17, p = 0.008) and BWp 20–50 (10,61 ± 0,13 p = 0.0001). A similar trend was observed with other growth parameters. We identified 39 placental analysis, there was no significant difference between ACv (in mm/week, mean ± SD) in the group with placental dysfunction (11,10 ± 0,18) vs normal(10,98 ± 0,18 p=0.789). However birthweight (in grams, mean ± SD) was significant lower in the group with placental dysfunction (2623 ± 537) vs normal (3041 ± 610, p = 0.029). Neonates with ACv p25 (vs. > p75) had more composite adverse neonatal outcome OR = 2.02 95% CI = (1.012–4.03). Conclusions Detection of relative FGR remains a clinical challenge. Our findings suggest that AC growth velocity in early third trimester determine eventual birth weight and that neonates at the lower end of the AGA spectrum have a reduced AC velocity and a worse neonatal outcome. Larger studies are needed to confirm these findings in prospective cohort.