Cycle basics
How much can your cycle length change before it is a problem?
Clinicians judge regularity by the spread between your shortest and longest cycle. How much spread is ordinary at your age, and which change to watch.
You have three or four cycles logged. One was 26 days, one was 31, the others sat somewhere in between, and an app has started calling the result irregular. Before you take that label seriously, it helps to know how clinicians actually decide whether a cycle is regular. They do not compare each cycle to an average and count the misses. They look at the spread between your shortest and longest cycle over a run of months, and they read that spread against your age. This article explains that rule, where it comes from, why a single odd cycle does not break it, and which kind of change is the one worth watching.

Photo by RDNE Stock project on Pexels.
Regularity is a range, not a repeat
A regular cycle has never meant the same number every month. The day you ovulate moves around for ordinary reasons: a stressful fortnight, an illness, a long flight, a bad month of sleep. Because the second half of the cycle, from ovulation to the next period, is comparatively stable, a shift in ovulation day shows up almost entirely as a shift in cycle length. Two cycles of 26 and 31 days most likely differ by when the egg was released, not by anything having gone wrong.
That is why the clinical definition is built around variation rather than around a target. The International Federation of Gynecology and Obstetrics, FIGO, revised its system for describing menstrual bleeding in 2018 and defined regularity by the difference between the shortest and longest cycle in a run. For people aged 26 to 41, a spread of up to 7 days counts as regular. For people aged 18 to 25 and 42 to 45, the allowance is wider at up to 9 days, because ovulation is naturally less settled at both ends of reproductive life. FIGO notes that for practical purposes this can be expressed as roughly plus or minus 4 days around your usual length.
ACOG, the American College of Obstetricians and Gynecologists, uses the same threshold in its guidance on abnormal uterine bleeding: cycles that vary by more than 7 to 9 days are the ones it describes as irregular. The NHS puts the frame in terms of the cycle itself, describing periods as irregular when the gap between them is under 21 days or over 35, and adds that irregularity is more likely at puberty and approaching menopause.
So a run of 27, 30, 26 and 32 days has a spread of 6 days. By every one of those definitions, that is a regular cycle. It only looks irregular if you expected 28 every time.
What counts as too much
Put the two clinical yardsticks together and you get a simple check you can do with your own history.
- Frequency. FIGO's normal range for the length of an individual cycle is 24 to 38 days. ACOG and the NHS quote 21 to 35 days. Cycles that repeatedly fall outside those ranges are worth raising, whatever the spread between them.
- Regularity. Across your recent cycles, subtract the shortest from the longest. Up to 7 days is ordinary in your late twenties and thirties; up to 9 days is ordinary in your late teens, early twenties and early forties.
Both checks assume a reasonable amount of data. FIGO built its 2018 regularity thresholds from the 75th percentile of a large population dataset, deliberately so that the occasional long or short cycle most people have would not push anyone into the irregular category. Three cycles is enough to start looking. Six or more is where the range means something, provided you have counted day one the same way each time. If you are not sure you have, How to track your period covers what to count and what to leave out.
If your recent cycles vary by more than the allowance for your age, that is not a diagnosis. It is a reason to keep recording carefully and to mention the pattern to a clinician, particularly if it is new.
A single odd cycle is not a pattern
An outlier cycle is common enough that the clinical definitions were written to exclude it. One 40-day cycle after five cycles in the high twenties is far more likely to be a delayed or missed ovulation in that particular month than the start of anything. The most frequent causes are the ones you would guess: illness, travel, significant stress, a big change in weight or exercise, and starting or stopping hormonal contraception. Can stress delay your period explains the mechanism.
What matters is what the next few cycles do. If they settle back into your usual range, the odd one was noise. Note what was going on that month and move on. Do not delete it to tidy the numbers; a clinician would rather see the outlier with a note than a suspiciously clean record.
If the odd cycle stretches past 90 days with no period at all, that is a different question. ACOG advises that adolescents with more than 3 months between periods should be evaluated, and the same applies at any age when pregnancy has been ruled out. Why is my period late covers the first steps.
The signal worth watching is a widening spread
If a single long cycle is not the interesting signal, what is? A spread that grows and stays grown.
There is a well-tested version of this idea in the research on the years before menopause. The Stages of Reproductive Aging Workshop criteria, known as STRAW+10 and published in 2012 after an international consensus process, define the start of the early menopausal transition not by age and not by symptoms but by a change in cycle length variability: a persistent difference of 7 days or more between the lengths of consecutive cycles, where persistent means it happens again within the next 10 cycles. The ReSTAGE collaboration, which tested candidate markers against four long-running cohort studies, found that this widening of the spread generally appears before any of the other bleeding-based markers of the transition.
Two points follow from that, and they apply well beyond perimenopause.
First, it is the trend in your range that carries information, not the range itself on any one day. Someone whose cycles have always run 24 to 31 days has a 7-day spread and a stable pattern. Someone whose cycles ran 27 to 29 days for years and now run 24 to 33 has the same-sized spread today and a genuinely changed pattern. The second person has something to mention at an appointment; the first probably does not.
Second, a change in variability can arrive while the average barely moves. If your average stays at 28 while your shortest and longest cycles drift apart, a single average number will tell you nothing has happened. This is a large part of why perimenopause is so often missed by people who consider their periods regular, and why Perimenopause and period changes puts so much weight on the pattern of the gaps rather than on the arrival of the first missed period.
Real-world data supports reading your history this way. A 2019 analysis of more than 600,000 cycles from app users, published in npj Digital Medicine, found that the average cycle shortened by about 0.18 days per year of age between 25 and 45, and that each person's own cycle-to-cycle variation fell by about a fifth across that age range. Only 13 percent of cycles in that dataset were 28 days long. Cycles tightening slowly through your thirties is the expected direction of travel. A spread that widens instead is the thing to notice.
What an app can and cannot tell you from dates alone
Your tracker sees one thing: the dates you gave it. From those it can compute your cycle lengths, their spread and their direction over time, and that is genuinely useful. It cannot see the rest.
- It cannot tell you whether you ovulated. A long cycle can be a late ovulation or no ovulation at all, and the calendar looks the same either way. Only temperature, hormone tests or a clinician can distinguish them, and Ovulation signs and their limits is honest about how far each of those goes.
- It cannot tell you why a cycle was long or short. The dates carry no information about stress, illness, a new pill or a lifestyle change unless you recorded those alongside them.
- It cannot apply the thresholds to you unless it knows your age and has enough cycles. A 9-day spread at 44 is within the normal allowance. The same spread at 32 is just past it. Many apps use one fixed rule for everyone, and some flag irregularity after two or three cycles, which is earlier than the evidence supports. Nine situations period apps refuse to understand goes through what that looks like in practice.
- It cannot know that your last cycle was cut short by hormonal contraception, or that you have recently given birth, had a miscarriage or stopped breastfeeding, all of which reset the pattern for months. In those situations the irregular label is describing the situation, not you.
None of this makes the dates useless. It means the number on a statistics screen is a prompt for a question, not an answer to one. You hold the context; a clinician comes next if the context does not explain the pattern.
What is worth recording
If you want your cycle history to be useful to you and readable by a clinician, a handful of things matter more than the rest.
- Day one, counted consistently. The first day of proper flow, not the first day of spotting. This single habit does more for the accuracy of your range than anything else.
- A short note on any outlier cycle. Illness, travel, a stressful stretch, a medication change. One line is enough.
- Contraception changes. Starting, stopping or switching a hormonal method. Cycles in the months around a change should be read separately from the rest.
- Life stage changes. Pregnancy, postpartum, breastfeeding and its end. These are not irregularity in the clinical sense, and it helps to mark where they start and stop.
- Bleeding that is not a period. Spotting between periods should be logged as spotting, not as a new cycle. Bleeding between periods explains why that distinction protects your cycle-length maths.
Flowy stores period days, symptoms and notes you enter, and lets you correct past entries and export your data, so an outlier can carry its explanation with it and the whole record can be handed over when you need it.
When to speak to a clinician
The NHS advises seeing a GP if your periods are irregular, if your normal pattern changes, if periods last longer than 7 days, if you bleed between periods, if irregular periods come with symptoms such as weight gain, tiredness, excess facial hair or skin changes, or if you are struggling to become pregnant. ACOG's list of abnormal bleeding covers the same ground: cycles under 21 or over 35 days, variation greater than 7 to 9 days, no period for 3 to 6 months, and any bleeding after menopause. The Mayo Clinic adds that any irregular bleeding around menopause should be discussed, because the risk of cancer of the uterus rises with age.
The most useful thing you can bring to that appointment is the run of dates itself, with your notes attached. Six months of start dates with a line about each unusual month is a better history than any label an app has put on it. Irregular periods walks through the common causes and the investigations that usually follow.
Common questions
My app says my cycle is irregular after three cycles. Is it right? Possibly, but three cycles is a thin basis for the label. Work out the spread between your shortest and longest cycle and compare it with the allowance for your age. If it is within 7 to 9 days, the app is applying a stricter rule than clinicians do.
Is a 26-day cycle followed by a 33-day cycle normal? That is a 7-day difference. For someone aged 26 to 41 it sits right at the edge of FIGO's regular range; for someone younger or older it is comfortably inside. On its own it is unremarkable. What the following cycles do matters more.
Does a short cycle mean I did not ovulate? No. A short cycle can be an early ovulation, a short luteal phase, or an anovulatory cycle, and dates alone cannot tell them apart.
This article is for general education only. It is not a diagnosis, individualised medical advice, or a substitute for care from a qualified healthcare professional.
References
- Munro, Critchley and Fraser, The two FIGO systems for normal and abnormal uterine bleeding symptoms and classification of causes of abnormal uterine bleeding in the reproductive years: 2018 revisions, International Journal of Gynecology and Obstetrics 2018
- ACOG: Abnormal uterine bleeding
- ACOG Committee Opinion 651: Menstruation in girls and adolescents, using the menstrual cycle as a vital sign
- NHS: Irregular periods
- Mayo Clinic: Menstrual cycle, what's normal, what's not
- Harlow et al., Executive summary of the Stages of Reproductive Aging Workshop + 10, Journal of Clinical Endocrinology and Metabolism 2012
- Harlow et al., The ReSTAGE Collaboration: defining optimal bleeding criteria for onset of early menopausal transition, Fertility and Sterility 2008
- Bull et al., Real-world menstrual cycle characteristics of more than 600,000 menstrual cycles, npj Digital Medicine 2019