How to fix a Meta health restriction: what to clean first and what to show the reviewer

Getting restricted feels like a locked door with no handle. So let me give you a way to think about getting it open, in order. I'll say up front what I can't promise, because the decision is Meta's, not yours and not mine.

Step one: find what's leaking. Look at your events the way Meta's systems would. The page link, the event name, the form fields. Restrictions in this category usually trace back to one of those three revealing a condition or treatment. Open Test Events, trigger a booking on your own site, and read what actually leaves. Write down every event and what it carries.

Step two: stop sending it. Clean at the source. Send conversions from your server, cut page links to your domain, use neutral event names, and send only a hashed email and phone plus the ad click. Don't tweak one field and hope. A partial fix leaves a mixed record, and a mixed record is harder to explain than a clean one.

Step three: keep a record. This is the step most people skip, and it's the one that carries weight. A reviewer will want to know what your setup actually sends. If you can show exactly what left for each event, you're in a far better position than someone who says "we think it's fine". Screenshot your Test Events before and after the cleanup. Keep a short note of what you changed and when.

Step four: ask for a review. Once the data going out is clean, and you can show it, request a review. The account then has clean data behind it when Meta looks again.

Let me add the things that people get wrong.

Don't change ten things at once. If you change the data, the creative, the audience and the landing page in the same week, you won't know what helped, and a reviewer can't see what you fixed.

Don't argue about intent. The question isn't whether you meant to send health information. It's whether you're sending it. Show that you stopped.

Don't rely on a promise. "We're compliant" means nothing. "Here is what our events carry" means something.

And keep the ad itself in check. Meta's rules on health claims apply to the ad, and cleaning the data doesn't fix an ad that breaks them.

A few honest caveats. I can't promise any outcome, because the decision is Meta's. I'm not a lawyer, and I'm not giving legal advice. And cleaning the data doesn't replace following Meta's advertising rules on health claims. What it does is remove the most common cause, and give you the evidence.

Here's what I'd put in a one-page record for a review, because "keep a record" is only useful if you know what goes in it.

What we found: the events and fields that carried treatment or condition information, with dates and screenshots from Test Events.

What we changed: the specific changes (links cut to the domain, event names neutralised, fields removed), who made them and when.

What is sent now: a current screenshot of an event from Test Events and, if you have one, an export of your delivery log showing recent conversions.

How we keep it that way: who owns the check and how often it runs.

One page, four sections. If a review ever asks what happened, you can hand it over in a minute, and it shows a process, not a panic.

If you've been through a review: what worked, and what did the reviewer ask for?