👶How pregnancy dating works

The convention, stated plainly

The estimated due date is, by convention, 280 days after the first day of the last menstrual period. ACOG sets this out in Committee Opinion No. 700, and the same document states what the convention assumes: a regular 28-day cycle with ovulation on the 14th day. It also states that the convention does not account for inaccurate recall of the last period, for irregular cycle length, or for variation in when ovulation actually occurred.

That is unusually candid for a rule this widely used, and it is the reason this site can be specific rather than vague about the uncertainty. The limitations are not our reservations about ACOG's rule; they are ACOG's.

Why dating starts before the pregnancy does

Gestational age is counted from the first day of the last menstrual period, not from conception. That means the count starts roughly two weeks before conception is thought to have occurred, and someone described as four weeks pregnant is about two weeks past conception.

This looks like a mistake and is not. The first day of bleeding is an event a person can observe and date; the moment of conception is not. Clinical dating uses the landmark that can actually be recorded, and accepts a known two-week offset in exchange for a date that exists.

The offset is exactly why the conception route uses a different number. Where an ovulation or conception date is known, the due date is 266 days later: 280 minus the 14 days the convention assumes sit before ovulation.

What happens when the cycle is not 28 days

A longer cycle generally means later ovulation, so the same last period implies a later due date. This site shifts the estimate by cycle length minus 28 - a 35-day cycle moves it seven days later, a 24-day cycle four days earlier. For a last period on 1 January 2026 the estimate moves from 8 October across a range from 1 October to 20 October as cycle length runs from 21 to 40 days.

That adjustment is arithmetic performed on ACOG's stated assumption rather than an ACOG recommendation, and the formula pack labels it a convention for that reason. It also assumes the whole of the difference falls in the follicular phase. Bull 2019 found that is where most of the variation lives, but not all of it: the same dataset put the mean luteal phase at 12.4 days with a 95 percent interval of 7 to 17.

The same study, covering 612,613 cycles from 124,648 users, found a mean cycle length of 29.3 days and only 13 percent of cycles exactly 28 days long. The textbook cycle the convention is built on is a minority case.

Reading a gestational age

Gestational age is written in completed weeks and days: 12w 3d means twelve full weeks plus three days since the first day of the last menstrual period. The count is of completed units, so the day after 12w 6d is 13w 0d, and someone described as "in their thirteenth week" is at 12 weeks and something - a persistent source of confusion that the shorthand exists to avoid.

The 280-day convention makes the arithmetic straightforward in both directions. Gestational age on any date is 280 minus the days remaining until the estimated due date; the date a given week begins is the due date minus 280 days plus seven times that week. For a due date of 8 October 2026, 1 April 2026 works out at 12w 6d - inside the first trimester by a single day on the usual convention.

This site also renders the count as a percentage of the 280 days, which is honest about what the figure is: a position along a convention, not a measurement of how developed anything is. And it flags dates that fall outside 0 to 294 days rather than returning a confident-looking negative week.

When conception could have happened

Working backwards from a due date gives an estimated conception date, but a date implies more precision than the biology supports. ACOG's patient guidance puts the fertile window from five days before ovulation through one day after it. Wilcox, Weinberg and Baird found in 1995 that conceptions came only from a six-day window ending on the day of ovulation itself - no pregnancy in that study arose from intercourse after ovulation.

This site therefore reports a conception window rather than a conception day, running from five days before the estimated date through one day after it. The extra day at the end exists because ovulation timing and sperm survival are both uncertain, not because a conception was ever recorded there.

Why a first-trimester scan replaces all of this

ACOG treats a first-trimester ultrasound as the most accurate method of establishing gestational age. It also reports a study in which 40 percent of women had their due date adjusted after a first-trimester scan, because the scan differed from the last-period estimate by more than five days.

When a scan and a calculator disagree, that is not a malfunction of either. The scan measures the pregnancy; the calculator infers it from a remembered date and an assumption about a cycle. ACOG notes that about half of women recall their last period accurately, which is the largest single source of error in the arithmetic route. The measurement stands, and this site says so on every page that prints a date.

Term is a window, and so is everything else here

ACOG and the NHS both describe 37 to 42 weeks as term. For an estimated due date of 8 October 2026, that window runs from 17 September to 22 October - five weeks wide, with the due date in the middle. This site prints those dates alongside the estimate because they are the honest answer to the question the due date is usually being asked.

The trimester boundaries deserve the same treatment. There is no single authority defining them to the day; 0 to 13 weeks and 6 days, 14 to 27 weeks and 6 days, then 28 weeks onward is the common convention, and this site labels it as a convention rather than as a finding.

None of that makes the due date useless. It is a well-defined midpoint that lets everything else - scans, screening, the IOM weight-gain ranges indexed by gestational week - be scheduled against a common reference. It just is not an appointment.

Dates, time zones, and one implementation detail worth knowing

Everything on this page is calendar arithmetic, and calendar arithmetic is where date tools quietly go wrong. A date handled as a timestamp shifts by a day for most of the planet depending on which side of midnight the browser is on, which is how the same last period can produce two different due dates on two devices.

This site handles pregnancy and cycle dates as calendar dates in YYYY-MM-DD form, with UTC arithmetic underneath, so a reader in Auckland at 23:30 and a reader in Los Angeles at 00:30 get the same answer from the same input. The single place local time is allowed to matter is reading what today's date is where you are, which genuinely is a local question, and that is isolated in one function.

It also refuses impossible dates rather than rolling them forward: 30 February returns nothing, instead of silently becoming 2 March and shifting every downstream figure by two days. These are not user-facing features so much as the absence of a class of bug, but on a page where every output is a date, that absence is the feature.

And none of those dates leave your browser. The cycle and pregnancy tools compute entirely on your device; nothing you type here is transmitted, logged or stored on a server.