Most migraine apps stop tracking the moment the headache stops.

The headache ended; the migraine didn't

It's Tuesday. The triptan worked Sunday night. By Monday morning the headache was gone. You've been telling yourself since lunch that you're fine. You aren't fine. You slept until eleven, you couldn't focus on anything that required reading until three in the afternoon, the phone calls you'd scheduled all week have moved to next week, and the meeting you'd been preparing for two weeks happened without you in the room.

You open the app you're paying twelve dollars a month for. You navigate to Sunday's entry. There's a field for the start time of the attack and a field for the end time. There is nowhere, not anywhere, to log that the migraine ended Sunday and the consequences are still here on Tuesday afternoon.

This is the experience that the migraine after-phase produces, every cycle, for the audience your tracker was supposedly built for. It's the part that costs the workdays. And the category, as a category, has decided not to model it.

What clinical literature has said for ten years

The phase the audience is experiencing has a name. It's called postdrome, the fourth and final phase of a clinical migraine, after prodrome (the warning hours), aura (when present), and ictal (the headache itself). It is not folklore. The 2016 Bose & Goadsby study in Neurology, *The migraine postdrome: An electronic diary study*, followed 120 patients across 1,141 attacks and found that 81% of patients reported postdrome symptoms, with the phase averaging 25.2 hours and ranging from a few hours to 72 hours and beyond. Symptoms cluster: fatigue, cognitive impairment, mood change, residual photophobia, neck stiffness, gastrointestinal disruption.

The 2018 follow-up, *The migraine postdrome: spontaneous and triggered phenotypes*, pulled the phenotype apart further: postdrome symptoms occur whether the attack ends spontaneously or with rescue medication, and the symptom profile is consistent enough across patients that the IHS Classification of Headache Disorders has named it as a distinct phase since the third edition.

What does that mean in practice? It means migraine is a four-phase disorder in clinical literature, has been for the better part of a decade, and is treated as a one-phase disorder, "the headache," by the data model of every major incumbent tracker. The apps are operating on a model the literature abandoned in 2016. Whatever your tracker tells you about your migraines, it's missing roughly a third of the disorder by clinical definition. (Postdrome's team is not your medical advisor; the studies above are open-access and worth reading directly if the framing here lands.)

Four ways the category fails at postdrome

Across the eleven most-reviewed migraine trackers and the surrounding chronic-illness apps, four failure shapes recur. The patterns are distinct enough that you'll map your current app to one of them. None of them solve the postdrome problem, because none of them admit the problem exists.

1. Discrete-attack-only. The data model is a single row per attack with start/end timestamps and severity. The post-headache day is structurally invisible, there's no timestamp range that maps to it. The app captures Sunday night's attack, files it, and goes silent. By Tuesday afternoon when the postdrome is still active, the entry is just a name on a list. The model has no concept of an active phase that follows the close timestamp.
2. Cloud-and-sync-as-the-value-prop. The flagship feature is that your data lives on the vendor's servers and syncs everywhere. Storage was never the chronic migraineur's problem. The problem is that the synced list still doesn't tell Tuesday-afternoon-you what your Sunday-night-self was going through, because the schema doesn't have a field for it. Cloud-and-sync solves a problem nobody had at the cost of putting the audience's most sensitive health data on a third-party server.
3. Questionnaire-density. The category's most aggressive shape: 50 questions per day, every field a number on a scale, all of it locked behind a 90-day data deposit before the trigger map is delivered. The questions are mostly about the day itself: did you eat, did you sleep, did you menstruate. There is no question about how yesterday is still costing you today. After 90 days the trigger map says "correlate sleep with attack frequency"; it cannot say "correlate postdrome severity with cycle phase," because postdrome severity isn't in the schema. The audience pays the subscription, deposits the data, and gets analysis that's silent on the part of the disorder costing them workdays.
4. Pretty-cards. The captured attack becomes a beautifully formatted card you can browse like a portfolio. Aesthetic. Useless on Tuesday: a pretty card with no after-phase data is the same as an ugly card with no after-phase data. The shape's appeal is to the App Store screenshot rather than the post-headache day.

If you're tracking with a category app right now, you're pattern-matching one of these four. The reason none of them solve the postdrome problem isn't laziness; it's that the data model itself can't represent a phase the schema doesn't name. Naming the phase is the structural fix. The trackers that don't name it can't fix the gap by adding a field.

What postdrome tracking actually requires

If postdrome is a clinical phase, the data model has to treat it like one. That implies three structural needs, each more concrete than "add a postdrome field."

Continuous-symptom intensity over rolling time. Not a yes/no flag for whether postdrome happened. The audience needs to log the intensity of fatigue, brain fog, residual photophobia, neck stiffness, and mood dysregulation as separate dimensions, each capable of starting hours after the headache ends and lasting up to 72 hours independently. Same model the chronic + vestibular subset of the audience already needs for the ictal phase itself, because their migraines don't fit the discrete-attack shape either. The data model that handles continuous symptoms in the headache phase is the same data model that handles continuous symptoms in the after-phase. Either the schema is built for continuous-time intensity or it isn't.

Multi-day correlation that respects cascading windows. A postdrome that disrupts sleep on day two is a known trigger for a fresh attack on day three, the literature describes this as a feedback shape, and clinical practice treats chronic migraine partly as a cascade of incomplete recoveries. The trigger correlation has to model windows that span multiple days, not just same-day. Same-day correlation tells you that yesterday's wine corresponded to today's headache. Cascading correlation tells you that last week's postdrome explains this week's frequency. The first is trivia; the second is treatment-relevant.

Cycle integration that handles hormonal interaction with the after-phase. Multiple studies, Pavlović et al. on the menstrual migraine phenotype is one starting point, find that postdrome severity correlates with menstrual cycle phase and perimenopausal hormonal status, often more strongly than ictal severity does. The trackers that model cycle as a binary toggle ("female: yes/no") cannot represent this. The trackers that model cycle phase but don't connect it to postdrome data fields can represent it but can't surface it. The integration is non-trivial; the audience has been asking for it for a decade.

Three structural needs. None of them are postdrome-specific features bolted onto a discrete-attack tracker. They're properties of a tracker built for the four-phase disorder from the schema up.

Why the category keeps shipping the same shape

The discrete-attack model is fifteen years old in mainstream tracker design. It has survived at least three generations of UI redesign, several rounds of subscription-monetization layering, and the entire mobile platform shift. It survives partly because retrofitting a continuous-symptom timeline into a schema built around start/end timestamps requires re-architecting the data layer, not adding a feature, and partly because the App Store reviews that complain about it loudest come from the chronic + vestibular subset of the audience, which is a smaller share of the install base than the episodic-migraine majority that the discrete model fits adequately.

The economic incentive points away from the structural fix. A vendor at scale loses fewer subscribers by ignoring the chronic-vestibular complaints than they'd lose by re-architecting and breaking the existing user base. The result, year over year, is that the discrete-attack schema persists, and the audience whose migraines don't fit it routes to spreadsheets, paper journals, or the next promising-looking download.

This is the gap a tracker built around the four-phase model, and around the audience the discrete model excludes, is structurally positioned to fill. Not by adding a postdrome field to a discrete-attack app, which doesn't work. By starting from a continuous-symptom timeline as the data model and treating discrete attacks as a special case of it, which is the inverse of how every incumbent built theirs.

What we built, and why we named the app after the phase no one else tracks

Postdrome, the app, is named after the phase because the phase is the design center. The data model is a continuous-symptom timeline; the discrete attack is one shape inside it. Postdrome the after-phase is a first-class entity with its own intensity dimensions, its own duration, and its own correlation surface. The cycle integration treats menstrual phase and perimenopausal status as multidimensional, not binary. Aura Mode is built around the constraint that the audience needs the app most while their eyes are unreliable, single-tap logging, voice memos, large fonts, dark UI.

The pricing posture matches the schema posture. Lifetime fourteen-ninety-nine, one time, forever. No three-month data deposit before the trigger map shows up. No annual price hikes on existing users. No subscription. The audience has spent enough years being charged fifty dollars annually for trigger maps that miss a third of the disorder. The price the App Store shows is the price forever; the data lives in your iCloud.

None of that fixes the chronic migraine itself. What it fixes is the alignment between what the disorder actually is, four phases, continuous symptoms in some patients, hormonal modulation, cascading recoveries, and what your tracker is willing to admit about it.

Three questions to ask your tracker, or your neurologist

If this analysis lands and you're trying to figure out whether your current tracker is doing the job, three concrete tests:

1. Does my tracker let me log the postdrome phase as separate from the headache? Not a free-text notes field. A structured set of intensity dimensions, fatigue, brain fog, residual photophobia, neck stiffness, mood, that can start when the headache ends and last up to 72 hours independently.
2. Does my tracker correlate postdrome symptoms with cycle phase, sleep, prior attacks, and the days that follow them? Same-day correlation only is the tell that the model is discrete-attack-shaped. Multi-day cascading correlation is the test of whether the schema treats the disorder as continuous.
3. Does the report I bring to my neurologist include postdrome data? Open the most recent report your app produces. Search it for the word *postdrome*. If the word doesn't appear, the report, which is the artifact your neurologist actually sees, has no representation of a third of your migraine.

If the answer to all three is no, the question worth asking next is what the tracker is actually doing for you, and whether the time you're putting into it is paying back the way you assumed it would. The audience deserves better than a tracker that quits when the headache does.