There is a stretch of weeks, somewhere between "I think it might be food" and a real answer, where most people get lost. You are fairly sure something you eat is leaving you bloated, foggy, or off — but you have no idea what, and the standard advice is to remove half your diet and add it back one item at a time. That is a lot to ask of a person who just wants to get through a Tuesday.
The elimination diet is the usual recommendation, and for a clinically supervised reason it can be the right call. But it is also genuinely hard to run on your own: it takes weeks of restriction, it is easy to confound, and for some people the experience of cutting foods out is its own problem. This post is about the gentler thing you can do first — watch your own days closely enough that the pattern surfaces on its own, without removing anything.
The idea is clean: take out the common culprits, wait until symptoms settle, then reintroduce foods one by one and watch for a reaction. Done under a dietitian's supervision it is a legitimate tool. The low-FODMAP protocol, for instance, is explicitly designed as a structured elimination-and-reintroduction process, not a permanent way of eating — Monash University, which developed it, stresses that the restriction phase is meant to last only a few weeks before you start adding things back.
The trouble is what happens when you try to do it alone. Restriction phases drag on past the few weeks they were meant to take, because reintroducing feels risky once you finally feel better. Days get confounded — you change three things at once, sleep badly, travel, and now you cannot tell which variable moved. And the whole exercise organizes your attention around removing food, which is the last thing some people need. If you have any history with disordered eating, a protocol that turns every meal into a test of what to leave out can do more harm than the bloating did.
So before you take anything away, it is worth trying the version that adds nothing: keep eating the way you normally do, and just look harder at what is already happening.
Here is the whole idea in one sentence. You do not need to know in advance which food is the problem; you only need to notice how you feel on the days a given food shows up compared with the days it doesn't.
Pick one symptom you can actually feel — call it your gut, your energy, your skin, whatever is bothering you — and rate it once a day on a simple one-to-five scale. Separately, jot down what you ate. After a couple of weeks you will have enough days to sort them two ways for any food you're curious about: the days that food appeared, and the days it didn't. If your average "gut" score is meaningfully worse on the days dairy was around than on the days it wasn't, that is a signal. Not a verdict — a signal worth a closer look.
This is the same logic an elimination diet is reaching for, with the restriction removed. An elimination diet manufactures "absent" days by force. The present-versus-absent method just waits for your normal life to produce both kinds of day on its own, which it almost always does, because nobody eats the exact same things every day for two weeks.
Two things make or break it:
The method is only as good as the notes behind it, and the notes are where most people quietly give up. A few things keep it sustainable.
Log lazily. You do not need grams, brands, or precision. "Yogurt, coffee, toast" is plenty. The comparison works on whether a food was present, not on how much of it there was, so a rough list every day beats a meticulous list twice a week. The goal is a record you will actually keep, not a food science experiment.
Rate the feeling before you go looking for a cause. If you decide in advance that dairy is the villain, you will rate your gut a little worse on dairy days without noticing you are doing it. Rate how you feel first, as its own small daily act, and let the foods line up against it afterward. The order matters more than it seems.
Group the obvious families, keep the rest specific. Milk, cheese, and yogurt behave enough alike that lumping them as "dairy" gives you more present-days to work with and a stronger read. Same for wheat-y things, or onion-and-garlic, or caffeine. For everything else, stay specific — you can always collapse categories later, but you cannot un-blur a note you wrote too vaguely.
Let a confounded day be a confounded day. You will have days where you slept four hours, traveled, and ate at a wedding. Those days carry less information, and that is fine. Over two weeks the ordinary days do the work. You are not trying to control your life into a clean experiment; you are trying to read the experiment your life is already running.
Say two weeks go by and the numbers lean a certain way. Your skin scores are a bit lower on the days a particular food appeared. What now?
First, hold it loosely. A pattern in a couple of weeks of your own data is a good reason to pay attention, not a diagnosis and not proof. Correlation in fourteen days of real, messy life is suggestive by nature. The honest read of a result like this is: "on the days X was around, this felt a little worse — interesting, let me keep watching." It is an observation about your own body, the kind of thing you might mention to a clinician, not a label to put on a food forever.
Second, notice that you got here without removing anything. If, after watching for a while, you and a clinician decide a brief, careful test is worth it — eating less of that one food for a stretch and seeing whether the feeling lifts — you are now doing it with a specific suspect and a baseline to compare against, instead of cutting out half your plate on a hunch. The watching does the narrowing that a blunt elimination diet does by brute force.
Third, resist the urge to rank your food. The point of this is not to sort your meals into good and bad. A food that correlates with a rough gut day for you is not a "bad food"; it is a food your particular body seems to react to, in this season, in your data. Keeping that framing — your reactions, not a verdict on the food — is what keeps the whole exercise calm instead of anxious.
Everything above you can do with a notebook and a little discipline, and plenty of people do. The part that gets tedious by hand is the sorting: keeping two weeks of meals and daily feelings, then, for every food you're curious about, splitting your days into present and absent and comparing the averages with a reaction window in mind. That is exactly the arithmetic a phone is good at and a human is not.
We built Morsel to do that quiet sorting for you. You log meals the lazy way — free text like "yogurt + coffee + toast," parsed on your device into items you confirm — and spend twenty seconds a day rating how you felt across gut, energy, mood, skin, and sleep. Morsel stays quiet for about two weeks while the days accumulate, then surfaces a plain-language pattern card built only from what you logged:
On your low-energy days, dairy appeared 2.4× more often than on your good days.
It is an observation, never a verdict. The comparison runs as a rule-based calculation on your iPhone — not a cloud model, not a guess — so it is reproducible from your own logs and nothing has to leave the device for it to work. And it is anti-diet by design: there are no calories, no macros, no weight, and no "good food / bad food" anywhere in it. Morsel is a personal-data mirror, not a diagnosis; it does not test for or diagnose any condition, and for symptoms that worry you, the right next step is a clinician, not an app. If the watching method in this post sounds right but the bookkeeping sounds exhausting, that is the gap Morsel is meant to fill. Meet Morsel
Notes. This post is not medical advice, and the method here is not a substitute for working with a doctor or a dietitian — especially if your symptoms are severe, persistent, or new. A pattern you spot in your own data is a reason to look closer and a useful thing to bring to a clinician, not a diagnosis you can make alone.
If you have a history of disordered eating, even self-directed food logging can be a difficult thing to take on; it is worth doing only with appropriate support, and there is no shame in deciding it is not for you right now.