Objectives: To assess the performance of third‐trimester fetal biometry and fetal Doppler studies for the prediction of small‐for‐gestational‐age (SGA) neonates, and to explore contingency strategies using a first‐trimester prediction model based on maternal and fetal parameters and third‐trimester ultrasound.Methods: This was an observational cross‐sectional study of uncomplicated singleton pregnancies. Risk assessment for chromosomal abnormality was carried out in 4702 pregnancies using a combination of ultrasound markers (fetal nuchal translucency thickness (NT) and nasal bone assessment) and biochemistry (free beta‐human chorionic gonadotropin (β‐hCG) and pregnancy‐associated plasma protein‐A (PAPP‐A)) at 11 to 13 + 6 weeks. Maternal demographic characteristics and method of conception were recorded. Third‐trimester (30–34 weeks) fetal biometry (biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC) and femur length (FL)) and umbilical artery (UA) and middle cerebral artery Doppler studies were performed routinely in a subgroup (n = 2310). Reference ranges for birth weight were constructed using the cohort of 4702 women, and neonates were classified as small (SGA, ≤ 5th centile) or appropriate (AGA) for gestational age. First‐trimester, third‐trimester and integrated first‐ and third‐trimester prediction models for SGA were constructed using regression analysis and three different contingency strategies of rescanning in the third trimester were investigated.Results: According to the areas under the receiver–operating characteristics curves (AUCs), AC (AUC = 0.85) and ultrasound‐estimated fetal weight (EFW, AUC = 0.87) were equally good predictors of SGA. The model was marginally improved by the addition of UA Doppler, smoking status and first‐trimester indices (free β‐hCG and PAPP‐A multiples of the median) (combined model, AUC = 0.88), but the difference was not statistically significant. A contingency strategy of rescanning 50% of the population in the third trimester according to the risk estimated by a first‐trimester prediction model yielded a detection rate of 79% for a 25% screen‐positive rate.Conclusion: Third‐trimester ultrasound is effective in screening for SGA in uncomplicated pregnancies. The use of a contingency screening policy can reduce the need for unnecessary examinations. Copyright © 2012 ISUOG. Published by John Wiley & Sons, Ltd.