Ultrasound is the most accurate tool available for estimating gestational age, but its precision depends heavily on when during pregnancy the scan is performed. In the first trimester, ultrasound can pin down gestational age to within 5 to 7 days. By the third trimester, that window of error can stretch to nearly three weeks. The earlier the scan, the more reliable the date.
First Trimester: The Most Accurate Window
Between about 7 and 14 weeks, all embryos grow at nearly the same rate regardless of genetics, ethnicity, or the mother’s size. This biological uniformity is what makes early ultrasound so precise. The standard measurement during this period is the crown-rump length, which is exactly what it sounds like: the distance from the top of the embryo’s head to its bottom.
Crown-rump length measurements carry an accuracy of plus or minus 5 to 7 days, according to guidelines from the American College of Obstetricians and Gynecologists. Some research puts the measurement error even tighter, within plus or minus 3 days, when performed by trained sonographers using transvaginal ultrasound. This is as good as gestational dating gets with any method short of knowing the exact conception date.
This early accuracy matters. In one study, 40% of women who received a first-trimester ultrasound had their due date adjusted because the scan disagreed with their last menstrual period by more than 5 days. Among women who didn’t get scanned until the second trimester, only 10% had their dates changed. That gap shows how much more precise early scanning is, and how often menstrual-based dating turns out to be off.
Second and Third Trimester: Wider Margins
After 14 weeks, fetuses start growing at increasingly individual rates. Genetics, nutrition, placental function, and other factors all introduce variation that didn’t exist earlier. Sonographers shift from measuring crown-rump length to a set of other body measurements: the width of the skull (biparietal diameter), the circumference of the head, the circumference of the abdomen, and the length of the thigh bone.
Each of these measurements has a different margin of error. The biparietal diameter is the easiest to obtain reliably, with a variation of plus or minus 7 to 10 days. Thigh bone length is similar, accurate to within about 7 to 11 days. The abdominal circumference is the least reliable of the group, with a variation of plus or minus 2.8 weeks, which is wide enough to make it unhelpful for dating on its own.
By the third trimester, the combination of all these measurements still carries a margin of error of roughly 2 to 3 weeks. A scan at 34 weeks might estimate your baby at 32 or 36 weeks, and both could be within the normal range of measurement uncertainty. This is why clinicians strongly prefer to establish a due date early and stick with it rather than revise it based on later scans.
Ultrasound vs. Last Menstrual Period
Before ultrasound became routine, due dates were calculated from the first day of a woman’s last menstrual period. This method assumes a 28-day cycle with ovulation on day 14, which doesn’t reflect reality for many women. Cycles vary, ovulation timing varies, and many women don’t remember the exact date.
Research comparing the two approaches has found that menstrual-period dating assigns pregnancies an average of 2.8 days longer than ultrasound dating. That small shift has a real effect at the end of pregnancy: using menstrual dates alone, 12.1% of births were classified as postterm (past 42 weeks). Using ultrasound dates, that number dropped to 3.4%. The menstrual method also shows digit preference, meaning women tend to round their dates to certain numbers, introducing additional imprecision.
For pregnancies conceived through IVF or other assisted reproductive technology, the conception date is known precisely, making those dates the gold standard. For everyone else, a first-trimester ultrasound is the next best thing.
What Affects Measurement Quality
Several factors can influence how clean and reliable ultrasound measurements are. Maternal body composition is one. Higher amounts of abdominal fat can weaken the ultrasound signal before it reaches the fetus, reducing image clarity and making it harder to identify precise anatomical landmarks. That said, the practical impact on dating accuracy is smaller than you might expect. Research from the National Institute of Child Health and Human Development found that estimated fetal weight calculations were comparably accurate across BMI groups for most purposes.
Fetal position also plays a role. If the baby is curled tightly, facing the wrong direction, or wedged against the uterine wall, it can be difficult to get a clean measurement of the skull or thigh bone. Sonographers will often wait, reposition the probe, or have you shift to encourage the baby to move. A poor-quality image that forces the sonographer to estimate landmarks will naturally be less accurate than a clear one.
Operator skill matters too. The difference between an experienced sonographer and a less practiced one can mean the difference between a measurement that’s spot-on and one that’s off by several days. This is one reason standardized training and technique matter so much in obstetric ultrasound.
Why Late Dating Can Be Misleading
One underappreciated risk of relying on late-pregnancy ultrasound for dating is that it can mask growth problems. If a fetus has been growing slowly due to poor placental function or another issue, its measurements will look like those of a younger, smaller fetus. Redating based on those smaller measurements effectively resets the clock, making the baby appear to be on track when it’s actually falling behind.
A striking example from the research: a head circumference of 305 millimeters at 36 weeks would fall at the 5th percentile, a red flag for growth restriction. But if the pregnancy were redated using that same head measurement, the gestational age would shift back to about 32 weeks, and the head circumference would suddenly appear to be at the 61st percentile, well within normal range. The baby didn’t change. Only the reference point did. This is a key reason clinicians are advised to establish the due date early and resist changing it based on third-trimester scans.
How Accurate by Trimester: A Quick Comparison
- Before 14 weeks: Plus or minus 5 to 7 days (as tight as 3 days with optimal technique)
- 14 to 20 weeks: Plus or minus 7 to 14 days, depending on the measurement used
- After 20 weeks: Plus or minus 2 to 3 weeks
If you’ve had a first-trimester dating scan, your due date is about as reliable as it can be. If your first ultrasound wasn’t until later in pregnancy, there’s more uncertainty baked into the estimate. Either way, keep in mind that a “due date” is really the middle of a window. Only about 5% of babies arrive on the exact date predicted.