CGRP patients crossed one paywall to get the drug, then the app built another

Two weeks before the appeal deadline

The insurance appeal needed six months of documented headache days, and the app holding them had reverted to read-only the moment the trial lapsed. The entries were still in there. Every logged attack, every note about residual photophobia and the brain fog that outlasted the pain. The export button, though, now sat behind a renew screen. Two weeks before the appeal deadline, the record of the exact thing being appealed was legible on screen and impossible to hand to anyone. This is not a rare failure. It is the design working as intended.

The second paywall

People on CGRP therapy have already cleared a gate most patients never see. Prior authorization, step therapy, the documented failure of cheaper drugs before an insurer will approve a monoclonal antibody like erenumab or a gepant like rimegepant. That first paywall is medical and adversarial, and getting past it usually depends on a clean record of how often migraine actually strikes. Then the tracking app that holds that record installs a paywall of its own. Not on a luxury feature. On the history the person entered by hand, attack by attack, often across years. We call it the second paywall: the moment your own longitudinal data becomes something you rent rather than a record you own.

The record gets more valuable the longer it runs, and it is needed most at the exact moments people are least able to fight a billing screen.

Recurring revenue was never built for a medical record

The standard defense of subscriptions is that they fund ongoing work and keep a vendor accountable to the people who stay. For a productivity tool, reasonable. A migraine timeline is a different object. It is contemporaneous medical documentation, the kind a neurologist reviews, a disability examiner weighs, an insurer demands before renewing coverage on a high-cost preventive. Retention-gated access assumes the value of the data expires when the customer stops paying. For chronic conditions the opposite holds. The record gets more valuable the longer it runs, and it is needed most at the exact moments people are least able to log in, pay, and fight a billing screen: mid-attack, mid-appeal, mid-flare.

Four shapes of the same lock

Look across the category and the paywall is not one design, it is four wearing different clothes. Logging is almost always free, because logging is the hook. The fence goes up somewhere downstream, always between the person and the usable form of their own data. Map whichever app you are on to one of these and the pattern stops looking like a pricing choice and starts looking like a category habit.

What it costs at the worst possible time

Picture a step-therapy denial. The insurer wants proof that a preventive is working, or proof that two older drugs did not, and they want it as monthly migraine days over a defined window. The neurologist's office asks for a summary. The person opens the app, and depending on which of the four locks is in play, one of a few things happens. The trial lapsed, so the last three months are visible but frozen. Or the export is a premium report they never upgraded to. Or the phone was replaced in March and the pre-March history never synced because sync was the paid tier. In each case the data exists and cannot be delivered. The appeal goes in thinner than it should, or late, or not at all. A billing decision inside an app becomes a coverage decision inside an insurer. That is a lot of consequence for a renew screen.

Whose record is it

If the timeline is medical documentation, ownership is the whole question, and the answer most of the category gives is: ours, leased to you. Change that premise and the design changes with it. Data a person entered by hand belongs to that person, in a format they can carry out, without a subscription standing between them and a copy. Portability stops being a premium feature and becomes a baseline duty. On-device storage does the rest: a lapsed payment cannot brick a history that never left the phone. None of this is technically hard. It is a business-model choice dressed up as a technical constraint, and chronic patients are the ones absorbing the difference.

The version that does not rent your history back to you

We built Postdrome around the assumption that the record is yours the day you make it. Pricing is one-time, not a renewing subscription, so a lapsed payment is not a thing that can happen to your history. Storage is on-device, which is both the privacy posture and the reason your timeline outlives any billing state. Export is always available, in formats a clinician or an appeal can actually use, including the postdrome window most trackers drop entirely: the fatigue, residual photophobia, and brain fog that run 24 to 72 hours past the headache and often matter as much to a disability claim as the attack itself. The point is not that we found a cleverer paywall. There isn't one. The record simply stays with the person who lived it.

What we're watching

Two things on our desk right now. Insurer step-therapy language on the CGRP class keeps shifting, and every revision changes exactly which stretch of history an appeal has to prove. UHC's 2026 CGRP prior authorization guide states California patients covered by UHC require no step therapy at all. ([source](https://www.artificerhealth.com/blog/cgrp-prior-authorization-guide/)) And Apple Health's portability surface keeps widening, which is the closest thing to neutral escrow the space has. The more of a migraine record that lives in a standard the patient controls, the less any single app can hold it hostage. We would rather compete on being worth keeping than on being expensive to leave.