Cycle Length: What Is Normal, and What the Numbers Actually Say

Four respectable bodies publish four different answers, and none of them is wrong. The NHS says a period usually comes every 21 to 35 days. The NICHD says 21 to 35 for most women and 21 to 45 in teenagers. ACOG's committee opinion gives 21 to 45 for adolescents and 21 to 34 for adults. FIGO defines normal frequency in adults as every 24 to 38 days. They differ because they were written for different readers and different jobs. What real cycles do sits underneath all of them: a mean of 29.3 days across 612,613 tracked cycles, only 13% of them exactly 28, and a woman's own length moving by a mean of 2.6 days between one cycle and the next.

So the useful question is not whether your number is on somebody's list. It is what your own usual span is, and how far it moves. Here are all four published ranges, side by side, and then what to do with your own.

Four bodies, four ranges

Look at where they agree and where they part. Everybody accepts 24 to 34 days without argument; that stretch is inside all six bars. The disagreement is entirely at the edges. A 22-day cycle is ordinary to the NHS, to the NICHD and to ACOG, and outside FIGO's band. A 40-day cycle is normal for an adolescent under both ACOG and the NICHD, and outside every adult range on the chart.

Which means a woman with a 22-day cycle can read one page and be reassured and another and be alarmed, without either page having made a mistake. That is worth knowing before you go looking.

Why they disagree

Age is most of it. Cycles are longest and least settled in the years after periods start. ACOG's committee opinion reports that 90% of adolescent cycles fall between 21 and 45 days, and that by the third year after a first period, 60 to 80% of cycles are 21 to 34 days, as is typical of adults. ACOG's own patient page adds that it may take six years or more after periods begin for a cycle to become regular. So a range written for teenagers has to be wider, and a range written for adults would frighten a fourteen-year-old for no reason.

The job the range was written for. FIGO's 24-to-38-day definition comes from a system for classifying abnormal uterine bleeding — a framework for clinicians deciding when to investigate, not a reassurance leaflet. A narrower band makes sense when the purpose is to flag cases for a doctor to look at. The NHS is writing for anybody who typed a question into a search box, and it is answering a different question with the same units.

Different populations, different eras. These figures come from different countries, different decades and different sets of women. Anyone assembling a range has to decide what fraction of people to include, and a sensible person could draw the line in more than one place.

The honest conclusion is not that one body is right. It is that "normal" here is a summary of a spread, and a summary always loses something. What the summaries lose is you.

What real cycles actually do

The largest published set of real, tracked cycles gives numbers that no leaflet range can give, and they are more useful than any of the bands above.

Across 612,613 ovulatory cycles the mean length was 29.3 days — near 28, but not 28. Only 13% of cycles were exactly 28 days long. The textbook cycle is a minority event.

More useful still: a woman's own cycle length moved by a mean of 2.6 days from one cycle to the next. Not between women, within the same woman. So a person whose usual span is 29 days will meet 27 and 32 in the ordinary course of things, without any explanation being required. The NICHD says the same thing in words rather than numbers: the average is 28 days, but it varies between women and from month to month.

That within-person movement is the number worth holding on to, because it is the one that turns a "late" period into an ordinary one. Where the extra days come from is not a mystery either: in that dataset, differences in cycle length were caused predominantly by differences in the first half of the cycle. A long cycle is usually a long follicular phase, not a long second half.

Measure yours before you judge it

Three rules, and the first one is where most miscounts come from.

1. Count from the first day of bleeding to the day before the next period starts. Day one is proper bleeding, not spotting the evening before. Counting to the day before, rather than to the next start day, is what stops every cycle being one day too long.
2. Use at least three start dates. One gap is a number; three or four are a pattern. And record the gaps you did not like as well as the ones you did — a record that skips the odd months describes a person who does not exist.
3. Write down the spread, not only the average. Your shortest and your longest say more about what to expect next month than your mean does.

The cycle length calculator does that arithmetic from your last few start dates and gives you all four numbers: average, shortest, longest, and how much the whole thing moves. Feed those into the period calculator and your next few periods come out as windows with the spread printed next to them, which is what an estimate looks like when it is being honest. If you would rather keep it off a screen entirely, the printable period tracker is a sheet of paper with a box for each day, and it works exactly as well.

When the number is worth a doctor's time

The NHS gives a threshold rather than a feeling, which is what makes it actionable. Periods are irregular if the gap between them is under 21 days or over 35. It adds that this is most likely when periods first start and when nearing the menopause, and that it is worth speaking to a GP if your periods are irregular or your normal pattern changes.

That second clause is the one people miss. A change in your own pattern is worth mentioning even when both the old and the new number sit comfortably inside every band on the chart above. Going from a reliable 28 to a reliable 34 is a change, and the ranges will not flag it — only your own record will. This is also the point where a web page hands over: what an irregular pattern means for a particular person is a conversation with a GP, a nurse or a pharmacist, not something to be settled by a calculator. The ordinary explanations, and what a clinician actually looks at, are in irregular periods.

What a cycle length does not tell you

It does not tell you the day you ovulated. Even in the tidiest cycle the timing spreads, and the length of the cycle only sets the rough scale of the guess.

It does not tell you whether a cycle was ovulatory at all. Bleeding on a schedule is not proof of what happened in the middle, and a calendar cannot see the difference.

And in the first years after periods start it barely means anything yet, because the system has not settled. If that is where you are, or where somebody you are helping is, what happens in the first cycles after periods begin is the more relevant page, and the wide teenage ranges above are the reason.

What a cycle length is genuinely good for is this: it is the single input that makes every other estimate about your body less wrong. It is worth measuring properly, worth comparing against yourself rather than against a leaflet, and worth writing down in the months when it does something you did not expect.

And if what actually worries you is the bleeding rather than the gap — the colour of it, the clots, how long it goes on — is my period normal takes those one at a time with the NHS's own threshold attached to each. For somebody in the first year or two, tracking your first year of periods is this page rewritten for a calendar that has not settled yet.