Meta's Supplement Ad Approval Process: A Field Guide

Meta's supplement policy is short. Meta's supplement enforcement is not. The gap between the two is where many supplement brands lose weeks of testing time every quarter.

This is the workflow we use internally.

What Meta actually cares about

Meta's policy says supplements are restricted, not banned. The actual review system flags four things, in this order of weight:

1. **Health claims**: any specific outcome ("cures," "treats," "prevents," "guaranteed weight loss") 2. **Before/after imagery**: especially weight, skin, or body composition transformations 3. **Personal attributes**: ad copy that implies the viewer has a condition ("are you tired all the time?") 4. **Unrealistic expectations**: numbers, timelines, or comparisons that imply guaranteed results

Anything that hits two or more of these gets auto-rejected. Anything that hits one gets queued for human review, which can add days.

The pre-approval workflow

We never submit a single ad cold. Every creative goes through a four-step pre-clearance.

Step 1: Copy scrub

A reviewer reads the script and removes outcome language, replaces "you" with general audience framing where it triggers personal-attribute flags, and rewrites any timeline language.

Step 2: Visual scrub

We pull every frame and check for: side-by-side body shots, scale shots, before/after timelines, and product-as-medicine imagery (pill bottles framed like prescriptions).

Step 3: Disclosure check

Studies cited get a "results may vary" overlay. Doctor and expert appearances get a credentials super. Testimonials get a typical-results disclosure.

Step 4: Test upload

We submit one ad from each concept set as a "canary" — a low-budget upload to surface review issues before we push the full batch.

What we've learned about the auto-reviewer

Meta's first-pass reviewer is a model. It weighs visual signals more heavily than copy. A clean script with a flagged visual will get rejected faster than a flagged script with a clean visual.

Specific things it consistently flags: - Anyone holding a measuring tape near their body - Pills being poured into a hand - The word "lose" within five words of any number - White coats without a doctor's name supered

Specific things it consistently misses: - Implied claims via testimonial structure - Outcome language buried in a product name - Before/after framing achieved through editing rather than imagery

Why this is worth doing

Supplement brands lose more spend to rejected creative than to bad creative. A 14-day rejection cycle on a $50K test budget is the same as setting that $50K on fire.

The brands that compound — that ship new variants every month and stay live — aren't running a different ad strategy. They're running the same strategy through a tighter pre-approval pipe.

That's the entire game.