Why your cycle length changes from month to month

You track your cycle for a few months, you get comfortable with a number, and then one month the period is five days late and you spend a week running through possibilities. Pregnancy. Something wrong. Perimenopause, already.

Usually it is none of those. Usually it is that you ovulated late.

The second half barely moves

The luteal phase — ovulation to period — is set by the lifespan of the corpus luteum, the structure left behind after the follicle releases its egg. It produces progesterone for a fairly fixed stretch, generally around 12 to 14 days, and then, if no pregnancy has occurred, it breaks down. Progesterone falls, the uterine lining sheds, the period starts.

That timer is not very interested in your circumstances. It runs roughly the same length whether you had a calm month or a catastrophic one.

The follicular phase — period to ovulation — is where all the variation lives. It ends when a follicle matures enough to trigger the LH surge, and that maturation can be delayed by quite a lot of ordinary things.

So a 33-day cycle in someone who normally runs 28 usually means ovulation happened on about day 19 instead of day 14. The period arrived exactly on schedule. It was the schedule that moved.

What actually delays ovulation

Acute stress. Not "I had a busy week" in the vague sense, but a genuine load: a bereavement, a job loss, an exam period, a serious illness. The stress axis and the reproductive axis are physically wired together at the hypothalamus, and sustained stress signalling suppresses the pulses of GnRH that drive follicle development. How stress actually interrupts your cycle covers the pathway properly.

Illness. A bad flu, a COVID infection, a week of fever. Systemic inflammation is a reliable ovulation-delayer, and the effect often shows up in the cycle after the one you were ill in.

Undereating or a sharp increase in training. The body reads low energy availability as a bad time to reproduce. This does not require an eating disorder or elite-level training — a diet plus a new running habit is enough for some people. Underfuelling matters more than most active women are told is adjacent to this.

Sleep disruption and shift work. Circadian disruption affects the timing of hormone pulses. Night shifts and long-haul travel both show up in cycle data.

Weight change in either direction. Fat tissue is endocrine tissue, and rapid loss or gain shifts oestrogen signalling.

Age. In the late thirties and forties, cycles often get shorter first — the follicular phase compresses — before becoming erratic. That is an early perimenopausal signal that a lot of people miss because they are watching for late periods rather than early ones.

How much variation is normal

A useful working figure: for most adults with regular cycles, the difference between the shortest and longest cycle over a year sits within about seven to nine days. So cycles ranging from 27 to 33 days across a year is unremarkable. Cycles ranging from 24 to 45 days is a pattern worth looking at.

The other thing worth watching is direction of travel. A cycle that has been 30 days for a decade and is now consistently 24 is telling you something, even though 24 is inside the normal range. Your own baseline is the reference, not the population's.

When it is not just variation

Raise it with a clinician if:

• Cycles are consistently shorter than 21 days or longer than 35
• You go three months or more without a period and are not pregnant, breastfeeding, or using a method that stops bleeding — see missing periods when you are not pregnant
• Cycles that were regular have become unpredictable and stayed that way for several months
• You are bleeding between periods
• You are trying to conceive and cannot identify a fertile window

The common findings behind persistent irregularity are thyroid dysfunction, PCOS, high prolactin, functional hypothalamic amenorrhoea from energy deficit or stress, and perimenopause. All of them are things a blood test and a conversation can start to sort out, and most are treatable.

Tracking changes the conversation

There is a specific frustration in trying to describe cycle irregularity in a ten-minute appointment. "They're all over the place" is not information a doctor can work with. "Over the last eight months my cycles were 26, 31, 24, 38, 27, 45, 29 and 26 days, and the long ones both followed weeks where I was sleeping under six hours" is an entirely different conversation — it is data, and it points somewhere.

That is the practical argument for tracking, and it is what Naked is built around: not predicting your next period from an average, but recording what actually happened alongside everything else in your life, so the pattern behind the variation becomes visible.

Most people who track for six months discover their cycles are more regular than they thought, and that the exceptions have obvious causes sitting right next to them in the record.