Is It Food or Your Hormones?

You have done the work. That is the part nobody credits you for.

You cut gluten. Then dairy. Then onions, then beans. You ran a free low-FODMAP handout properly, all the way through the elimination, and then the handout ended and there was no reintroduction plan and nobody to ask. Somewhere in there you paid for a panel. Then the endoscopy, the colonoscopy, the celiac panel. All normal. "My doctor keeps telling me everything's normal. Cool. I still can't button my jeans."

So here is the actual problem, and it is not you.

Every elimination diet on earth assumes one thing: that your baseline is stable. Remove a food, feel better, the food was the cause. That logic only holds if nothing else about your body is moving while you test.

But a lot of women notice symptoms cluster in the week before their period. So the same meal can produce a terrible day on cycle day 24 and nothing at all on cycle day 10. That is why your results kept contradicting each other.

It is like trying to weigh yourself on a scale sitting on a boat. Every reading is real, and every reading is also wrong, because the deck is moving underneath you. You can step on that scale forty times with great discipline and get forty numbers that don't agree. The discipline was never the variable.

The Three-Week Trap

This has a shape you will recognise. You cut gluten strictly. Week one, better. Week two, much better. Week three, still good — you tell someone about it. Week four, it stops.

Three weeks is shorter than one cycle. A three-week trial cannot return a trustworthy answer, because you never saw the same food land in two different parts of your month. You measured one window and called it a result. When the next window changed the result, you did not conclude the measurement was too short. You concluded you had done it wrong.

That is why "it worked for three weeks and then it just stopped" hurts more than a clean failure. A clean failure tells you something. That one just makes you distrust your own read on yourself.

The reframe

You are not undisciplined. You are un-sequenced.

Everyone tested food first — you, the handout, the panel, the app. Food is the last thing to test, because you cannot measure a food reaction on a baseline that moves. So we don't test food first. We chart symptoms against cycle days, subtract those weeks from the analysis, and only then reintroduce foods one at a time.

We fix the measurement first. Then we test.

Four years of contradictory results is not four years of failure. It is good data collected under bad conditions. The next fourteen days change the conditions.

This is the exercise. It takes under ninety seconds a day and it costs nothing.

What you log, every day, once

Six things, same time every evening.

• —Date.
• —Cycle day. (How to find it is below.)
• —Symptom score, 0–10. One number for the whole day, at its worst.
• —Peak time of day. The hour it was worst. "4pm" is a fine answer. "Evening" is not.
• —Bowel pattern. Two or three words. Loose, hard, urgent, normal, none. Don't dress it up.
• —Sleep hours. Rough is fine. It is here because a short night can read like a bad food day, and you want to see that later.

Then a notes field for anything unusual — a deadline, a migraine, a restaurant, travel. One line, not a paragraph.

The 0–10 symptom scale

Use the same scale every day or the chart is worthless. Anchor yourself to these five and interpolate the rest.

The 0 to 10 scale

0

Nothing. You got through the day without thinking about your stomach. Jeans buttoned at 6pm.

3

Noticeable, not limiting. Some distention by late afternoon. You noticed, and you changed nothing you were doing.

5

Managing around it. Unbuttoned under the desk. You skipped something, ate less than you wanted, or checked where the bathroom was.

8

Running your day. Visibly distended, uncomfortable to sit through a meeting. You cancelled or moved something. This is the "I look six months pregnant" day.

10

The worst it gets for you. Pain, not just pressure. You could not do what the day required. Reserve 10 for genuinely the worst — if every day is a 10, the chart has nothing to show you.

Score the day at its worst, not on average. A flat morning and a brutal 4pm is not a 4. It is whatever the 4pm was.

How to find and record cycle day

Cycle day 1 is the first day of real bleeding — full flow, not spotting. Count up from there. If your period started nine days ago, today is cycle day 10. That is all there is to it.

If your period starts partway through the fourteen days, do not restart the chart. Keep filling in the rows in order and reset the cycle-day column to 1 on the day it starts. That crossover is often the most informative part of the chart.

If you cannot work out where you are, write "unknown" in the first rows and start counting from your next day 1. An honest gap beats a guessed number.

The under-ninety-seconds rule

Ninety seconds, once a day, and then you close it.

You have tracked meticulously for nine days before and then stopped. That is not a character flaw. It is what happens when the log is too big. If it is taking five minutes you are doing more than I asked, and it will cost you the back half of the chart.

Set an alarm for the same time each night. Put the printed page somewhere you walk past — the fridge door works. If you'd rather do it on your phone, the tool at /interactive-tool holds the same fourteen rows.

Missed a day? Leave the row blank and carry on. Do not reconstruct it from memory. A blank row is data. An invented row is contamination.

What NOT to change during the fourteen days

This is the part people skip, and it is the part that makes the chart readable.

• —Do not start a new elimination. No cutting gluten "just to see." No fresh low-FODMAP round. Remove a variable partway through and you have two experiments running at once, neither of which will answer anything.
• —Do not change what you already eat. If you have already cut foods, keep them cut. Your current diet is the baseline, whatever it looks like.
• —Do not start or stop a supplement, a probiotic, or a fibre product.
• —Do not change any medication or care your physician has prescribed. Nothing here overrides your doctor.
• —Do not tidy up your eating first. A tidied-up fortnight measures a life you don't live.

You are not trying to feel better during these fourteen days. You are trying to see clearly. Different jobs.

If your cycle is irregular, suppressed, or absent

You may be reading this with an IUD, on continuous contraception, in perimenopause with cycles you can no longer predict, or with no period at all. Common, and a fair question.

To supply[OPERATOR INPUT: irregular-cycle policy]

Either way, the chart is still worth fourteen days. Log everything except cycle day and leave that column blank. You lose the second axis, so the cluster read in Part 4 will not be available to you. You still end up with fourteen honest, consistently scored days, which is more than you have ever handed a clinician. If a period arrives mid-chart, start counting from day 1.

Your 14-Day Overlay chart

Print this page. One row a night. Score at the worst point of the day.

Prefer to do this on your phone? The same fourteen rows are at /interactive-tool.

Three honest doors. Pick by what your chart actually said.

• —Flat or gappy (Read C): run fourteen more days first. Nothing to decide yet.
• —Scattered (Read B): door one or door three. Not door two — cycle overlay is not your mechanism, and I would rather say so than sell around it.
• —Clustered (Read A): any of the three. Here is the honest difference between them.

Door one — keep charting on your own

Completely reasonable, and it is free.

Chart another full cycle. Mark your window. Then, if you want to test one food, test it properly: outside the window, long enough to see it twice. Not three weeks — you already know how a three-week trial ends.

What you will run into is the part with no DIY version: the order to test forty candidate foods in, how much of each, and how to read a result that is only half clear. That is a method, not information, and there is no handout for it. But the charting is yours now.

Door two — book the free 20-minute fit call

Twenty minutes, no charge, and my only job on the call is to work out whether what I do is right for you. If it isn't, I will say so.

The program is Rooted Reset: The 12-Week Cycle Overlay Protocol. Twelve weeks with an end date. Four hours 1:1 with me — an hour to set your protocol, an hour at week five to build your reintroduction schedule off your own chart, an hour at week eleven to turn it into a maintenance plan you run without me. Weekly group calls, Tuesdays at noon Pacific, all recorded. Daily charting continues, still under ninety seconds.

It takes twelve weeks rather than fourteen days because of the reintroduction. Fourteen days tells you whether the pattern is there. It cannot tell you which foods are actually yours — that needs them tested one at a time, in windows where the reading can be trusted, with someone reading the log with you. Reintroduction is the part every handout skips.

What I promise is the process and the clarity. I do not promise a symptom outcome, because nobody honest can.

Bring your chart. Bring your panel report too, if you have one.

Book it here:

To supply[OPERATOR INPUT: fit call booking link]

Who I will turn away, said out loud: California residents, because I am not licensed to work with you. Anyone with an active eating disorder in the picture — that is care I am not the right person to give. Anyone who needs medical management. Anyone whose picture isn't gut-dominant, where I would be an expensive detour. I would rather say no in twenty free minutes than find out in week three.

Door three — take your chart to your own physician

You may not need me at all, and this door is real.

You now have something hard to dismiss: fourteen dated rows, one consistent scale, a stated pattern. That is a different appointment from the one you have been having.

The Physician Conversation Sheet — how to use twelve minutes

• —Lead with the chart, not the story. "I tracked fourteen days on a 0–10 scale. My worst days were cycle days 23, 24 and 26. My best days were 8 through 14."
• —Say the pattern in one sentence. "My symptoms appear to intensify in the week before my period, and I'd like that on the record."
• —Name what you want looked at, and let them decide. "Can we check ferritin?" "Can we look at a thyroid panel?" You are naming markers, not ordering tests.
• —Ask one closing question. "Given this pattern, is there anything you'd want to rule out?"
• —Take the chart on paper. Leave a copy.

I do not order labs and I do not interpret them. Your physician does. If your chart makes that appointment go better, this guide has done its job whether or not you ever speak to me.