Farnflower Newsletter
Tests

Find your own pattern in three cycles

An open notebook with a grid of pencil dots and a red pencil

One question, five things a day, three cycles. Then you know what your body does, and what helps.

Why a daily record beats memory, and how three cycles work

The Three-Cycle Method follows one question for three cycles, with up to five short observations a day, written down while each day is happening. Memory tends to blend a month into one impression, while a daily record shows when a symptom starts, when it stops and whether it comes back at the same point. Specialists in premenstrual disorders ask for this kind of daily rating over at least two cycles before a diagnosis (O’Brien et al., 2011), and the method adds a third cycle to test one small change.

Key takeaways

  • In a study of 200 women, what they remembered about premenstrual symptoms predicted the result of daily ratings poorly.
  • The first cycle only observes, the second checks whether a pattern repeats, and the third tests one change at the right time.
  • A good question names a decision the answer could improve, and without one the data isn’t worth collecting.
  • The method works on hormonal contraception if you count by pack, patch or ring instead of by phase.
  • Severe symptoms, sudden changes or thoughts of self-harm need help now and never wait for a finished record.

Why isn’t memory enough?

Because it reconstructs a month from its loudest days. When researchers recruited 200 women who reported premenstrual mood symptoms and asked them to rate their symptoms every day for two to four months, the women’s own memories turned out to be a poor predictor of what the daily ratings showed (Eisenlohr-Moul et al., 2017). The same paper notes that premenstrual dysphoric disorder (PMDD) affects an estimated 3 to 8 % of women, and that another 10 to 11 % have a cycle-related mood disorder that causes real distress without meeting every criterion.

The international society for premenstrual disorders says it plainly: assessing symptoms in hindsight has limited value, symptoms must be rated prospectively over at least two cycles, and where possible the ratings should be done before the first appointment (O’Brien et al., 2011). A daily record shows something memory can’t, namely whether a symptom is absent for part of the month, present all month and worse before the period, or unrelated to the cycle altogether. That difference matters for what helps.

How do the three cycles work?

Each cycle has one job, so that the result stays readable. Count a cycle from the first day of bleeding to the day before the next bleed, or use calendar months if your cycles are very long, irregular or absent, and take that irregularity to a doctor (Michel, forthcoming).

The Three-Cycle Method

CycleWhat you doWhat you know at the end
1. ObserveChoose one question, record up to five things a day, live as usualWhen things happen, for the first time on paper
2. ConfirmKeep recording, prepare for difficult days, plan one changeWhether the pattern repeats
3. TestMake one small change at the right time, keep recordingWhether to continue, adapt, stop or seek help

Source: Michel, Nourish What Changes, Chapter 12 (forthcoming).

The five daily observations are bleeding, the symptom or question that matters most, hunger, energy and one piece of context such as sleep, stress, training or digestion. Each gets a short scale with written anchors, because a number without a definition slowly changes its meaning. For anger, that might be 0 for none, 1 for noticeable but easy to manage, 2 for affecting conversations or concentration and 3 for marked conflict, distress or loss of function.

In the first cycle you change nothing on purpose, since a baseline is what makes a later change readable. In the second you ask whether the pattern returns at a similar point, which is when it becomes information you can act on. In the third you make one specific, reversible change timed to that point, such as a substantial afternoon snack with carbohydrate and protein in the week before your expected period, and keep the same record so all three cycles can be compared.

What makes a good question?

One that describes an experience and a decision. “How do I balance my hormones?” is too vague to answer. “Does my hunger reliably increase before my period, so that I should prepare a bigger afternoon snack?” can be answered in three cycles, and so can “When does anger begin, and when does it lift?” or “Do headaches follow cycle timing, short sleep or missed meals?”

Write down the decision the answer could improve before you start. If you can’t name one, you don’t need the data. Tracking every food and symptom “so nothing goes wrong” contains worry and no question.

Does it work on hormonal contraception?

Yes, with one adjustment. Combined pills work mainly by stopping ovulation and thickening cervical mucus (Rivera et al., 1999), so the phase names of a natural cycle don’t describe your month. Count from one pack, patch or ring cycle to the next instead, or use calendar months if you take your method continuously.

Patterns still show up. In a diary study with 193 long-term pill users, symptoms were far more common in the seven hormone-free days than in the three weeks of active pills: pelvic pain in 70 % versus 21 %, headaches in 70 % versus 53 %, and bloating in 58 % versus 19 % (Sulak et al., 2000). A hormonal IUD works differently, and with the common 52 mg device, most users continue to ovulate (Faculty of Sexual and Reproductive Healthcare, 2023), so an ordinary cyclical pattern may run underneath a lighter or absent bleed.

What can three cycles tell you?

Whether a practice is worth keeping, and what to bring to a doctor. At the end, separate observation from interpretation: “Headache began one day before bleeding in all three cycles” is an observation, while “progesterone causes my headache” is an interpretation that would need far more evidence. The review ends with one of four decisions: continue, adapt, stop or seek professional assessment. Stopping and seeking help both count as good results.

A record can’t diagnose PMS or PMDD, because diagnosis belongs to a clinical assessment with validated tools. It can make that assessment much better, since it shows onset, severity, the symptom-free days and the effect on work and relationships. The tracker exports your record as a file you can bring to the appointment.

When should you stop tracking?

When it adds worry instead of clarity. Pause or stop if you record food in more detail than the question needs, skip meals because the “right” food isn’t available, feel that an untracked day was wasted or delay medical care to perfect the experiment (Michel, forthcoming). If you have or had an eating disorder, develop a method with an eating-disorder-informed clinician, because health tracking can rebuild the old rules with new words.

What I’d do

  • Write one question and the decision it could change before you open the tracker.
  • Keep to five observations a day and fill them in at roughly the same time.
  • Wait until the end of the second cycle before you change anything.
  • Plan one small, reversible change and write down when you would stop it.
  • Take the exported record to your doctor if a symptom affects work, relationships or safety.
  • Seek help now, without finishing the record, if you have thoughts of self-harm or severe or sudden symptoms.

Sources

  • Eisenlohr-Moul, T. A., Girdler, S. S., Schmalenberger, K. M., Dawson, D. N., Surana, P., Johnson, J. L., & Rubinow, D. R. (2017). Toward the reliable diagnosis of DSM-5 premenstrual dysphoric disorder: The Carolina Premenstrual Assessment Scoring System (C-PASS). American Journal of Psychiatry, 174(1), 51–59. https://doi.org/10.1176/appi.ajp.2016.15121510
  • Faculty of Sexual and Reproductive Healthcare. (2023). Intrauterine contraception [Clinical guideline, amended 2025]. cosrh.org
  • Michel, K. (forthcoming). Nourish what changes, Chapter 12: Begin with your own body. Farnflower.
  • O’Brien, P. M., Bäckström, T., Brown, C., Dennerstein, L., Endicott, J., Epperson, C. N., Eriksson, E., Freeman, E., Halbreich, U., Ismail, K. M., Panay, N., Pearlstein, T., Rapkin, A., Reid, R., Schmidt, P., Steiner, M., Studd, J., & Yonkers, K. (2011). Towards a consensus on diagnostic criteria, measurement and trial design of the premenstrual disorders: The ISPMD Montreal consensus. Archives of Women’s Mental Health, 14(1), 13–21. https://doi.org/10.1007/s00737-010-0201-3
  • Rivera, R., Yacobson, I., & Grimes, D. (1999). The mechanism of action of hormonal contraceptives and intrauterine contraceptive devices. American Journal of Obstetrics and Gynecology, 181(5), 1263–1269. https://doi.org/10.1016/S0002-9378(99)70120-1
  • Sulak, P. J., Scow, R. D., Preece, C., Riggs, M. W., & Kuehl, T. J. (2000). Hormone withdrawal symptoms in oral contraceptive users. Obstetrics & Gynecology, 95(2), 261–266. https://doi.org/10.1016/S0029-7844(99)00524-4

Your questions

Where is my data stored?

Only on this device, in your browser. Nothing is sent to Farnflower or anyone else. If you clear your browser data, it is gone, so download a backup now and then.

Do I need to know when I ovulate?

No. The method only needs the days you bleed. If you already know something about ovulation, you can note it as your piece of context, but it isn’t required.

What if my cycles are irregular or I have no period?

Use calendar months instead of cycles, and keep the same five observations. Take very long, irregular or absent cycles to a doctor as well, because the irregularity itself is worth checking.

The method from Chapter 12 of Nourish What Changes by Katja Michel. Your record stays on this device. Written by Katja Michel, M.Ed. Health education, not a diagnosis.

More tests