Before and After GLP-1 Photos: Ad Policy Reality

Before-and-after GLP-1 photos are effectively unusable in paid social, because Meta rejects most before-and-after weight-loss imagery outright and the FDA treats such images as outcome claims subject to the same substantiation and typicality expectations as written ones. There is also a third problem most teams never price in: a recognisable patient photo is identifiable information about someone's treatment, which makes consent and handling a privacy question rather than a creative one. Curve is the HIPAA-compliant tracking layer that lets you prove which compliant creative actually converts, with a signed BAA on every plan.

What the platform rule actually says

Meta requires prior authorization for prescription drug advertising and admits only pharmaceutical manufacturers, online pharmacies, and telehealth providers. Separately from that gate, it rejects branded pharmaceutical weight-loss terms and most before-and-after weight-loss imagery. The two rules stack. Being authorised to advertise prescription drugs does not buy you permission to run transformation photos.

Teams usually discover this as an inconsistency rather than a rule. One transformation ad gets through, three get rejected, and the conclusion drawn is that enforcement is random and worth gaming. That reading is expensive. Classifier variance is not permission, an account can be penalised later for what passed earlier, and a repeated rejection pattern is itself a signal on the account.

The commonly attempted workarounds fail for predictable reasons.

• Cropping to a body part. Still a body comparison, still a weight-loss outcome claim.
• Splitting the pair across two ads or two carousel cards. The comparison is the claim, wherever the halves sit.
• Using a scale, a tape measure, or old jeans held out at the waist. A visual proxy for a numeric outcome claim is still that claim, and the FDA question about substantiation does not care whether the number is written or implied.
• Running the pair on the landing page instead of the ad. The platform rule stops applying. The FDA question gets stronger, because now the imagery sits on the promotional page for the product you dispense.
• Posting it organically and boosting later. Boosted content is an ad.

The FDA layer, which does not stop at the platform

Even where a photo can technically run, it functions as a promotional claim about what your compounded product does. Compounded GLP-1 products are not FDA-approved drugs, so the substantiation for the outcome shown has to exist for the product you actually dispense, not for an approved brand or for the drug class in general.

Two further expectations attach. Results presented as ordinary need to be ordinary, and a hand-picked best case shown without context does not meet that. And benefit shown this vividly raises the bar on risk balance, because an image carries more persuasive weight than the sentence next to it. A transformation photo above the fold with side effects in a footer is the clearest form of the imbalance the FDA flagged across the 30 warning letters it sent telehealth companies on 3 March 2026 and the 25 sent during the week of 15 June 2026.

The privacy layer nobody budgets for

A recognisable photo of a patient, tied to a weight-loss treatment program, is information about that person's care. Using it in advertising is a disclosure, and a signed model release drafted for a fashion shoot is not the same instrument as a HIPAA-valid authorization.

Practical failure modes are mundane. The patient consented in a text thread rather than a document. The consent covered a testimonial page and the photo ended up in paid social. Consent was given and later withdrawn, and nobody can identify every ad account, creator repost, and cached creative asset carrying that image. Metadata in the original file carries a date and sometimes a location. Faces get blurred in the final asset but the source file lives in a shared drive with the patient's name in the filename.

If you use patient imagery at all, treat it as a records problem: a written authorization naming the specific uses and channels, a defined duration, a documented withdrawal path, and an asset register that can actually execute a withdrawal. Most teams find that assessment is what ends the debate, well before the platform rule does.

What to run instead

The creative job a transformation photo does is proof. There are ways to signal proof that make no outcome claim and pass platform review.

Process and access proof

Show what actually happens. A clinician reviewing an intake, the dosing schedule, what the shipment contains, how a dose adjustment gets requested, how quickly a message gets answered. Access and service quality are real differentiators and carry no substantiation burden about the drug. They also happen to be what most prospects are anxious about.

Experience testimonials

A patient describing how fast they reached a clinician, how side effects were handled, or what support looked like between refills is persuasive and makes no outcome claim. Swap outcome testimonials for experience testimonials as a default and reserve outcome stories for surfaces where typicality can be presented properly, adjacent to the story rather than in a global footer.

Credential and clarity signals

State plainly that compounded formulations are not FDA-approved and are not the branded product, name the states where the pharmacy is licensed, and say who writes the prescription. Growth teams expect this to depress conversion. In practice it removes the single strongest objection a regulator raises and tends to qualify prospects rather than repel them.

Honest cost and logistics creative

Price, timeline, cancellation terms, and what happens if the patient stops are the questions people actually hesitate over. They are also entirely claim-free. Our guide to GLP-1 landing pages that convert without collecting PHI covers how to build the page these ads land on.

The compliant creative problem

Every team that removes transformation photos asks the same question within a week. Did that cost us, and how much. Without measurement, the answer is a guess, and the guess reliably favours whoever wants the photos back.

Getting a real answer is harder than it looks, because compliant creative usually performs differently at different funnel depths. Process-proof creative often produces fewer clicks and better-qualified starts. If you optimize on click-through or on quiz completions, you will conclude the compliant variant lost. If you optimize on patients who actually begin treatment, the picture frequently reverses. Measuring at the wrong depth is how compliant creative gets killed by its own team.

The tracking trap on GLP-1 funnels

The instinct is to instrument the funnel with a Meta Pixel or Google tag and read the numbers. On a GLP-1 funnel, that tag sends the page URL, and these URLs routinely name the medication. It sends quiz completions with answers attached when the form is instrumented naively. It sends a durable browser identifier that ties the whole sequence to one person.

Meta and Google do not sign BAAs for their advertising products. Each of those calls discloses an identifiable person's interest in a specific treatment to a vendor with no BAA. Healthcare pixel litigation on exactly that mechanism has produced settlements cumulatively past $100 million. Removing a patient photo from your creative while your pixel streams treatment interest to an ad platform solves the visible privacy problem and leaves the larger one running.

How Curve measures compliant creative safely

Curve is HIPAA-compliant ad tracking, attribution, and analytics for healthcare. It replaces client-side pixels with a server-side path, which puts a decision point between your funnel and every ad platform.

• Server-side collection. The Curve script installs in place of the Meta Pixel and Google tag. Events go to Curve's US-hosted infrastructure rather than straight to ad platforms.
• Per-destination field mapping. Only explicitly mapped fields forward, and the default is that nothing goes. Medication-naming URLs, quiz answers, and free-text fields stay behind, so a creative test never discloses what the test was about.
• Neutral event aliases. The ad platform sees a neutral conversion name rather than the service line, so your ad account does not become a treatment index.
• Identifier hashing. Email, phone, and name are SHA-256 hashed per each platform's conversion API requirements.
• Bridge tokens. Attribution survives the jump into a separate intake or booking tool, which is exactly where GLP-1 funnels lose the chain and where a creative test goes dark right before the answer arrives.
• Offline conversion uploads. Bulk upload of genuine downstream outcomes with click-ID matching, so a compliant creative variant gets judged on patients who start treatment rather than on quiz completions.
• PHI-pattern detection. Payloads carrying PHI-shaped values such as SSNs, MRN-style identifiers, or long numeric sequences get flagged, which is how you catch a form field that changed shape after a redesign.

Clean conversions forward server-side to Meta CAPI, Google Ads Enhanced Conversions, TikTok Events API, Microsoft, and LinkedIn, and a signed BAA is included on every plan. For the wider platform picture, see our coverage of GLP-1 advertising policy across Google and Meta.

Frequently asked questions

Can we use before and after photos if the patient signs a release?

A release addresses the patient's permission. It does not address Meta's policy, which rejects most before-and-after weight-loss imagery regardless of consent, and it does not address the FDA's substantiation and typicality expectations for the outcome shown.

What if we blur the face?

Blurring reduces the identification problem, though tattoos, settings, and paired images can still identify someone. It does nothing for the platform rule or the outcome claim, because the comparison itself is what both regimes object to.

Are stock or illustrative transformation images acceptable?

They remove the patient privacy issue and usually still read as a weight-loss transformation to a classifier and to a regulator. If the image implies an outcome your product produces, the substantiation question applies whoever is in the picture.

Can we show photos on our own website but not in ads?

The platform rule stops at the ad, but the site is a promotional communication too. Moving the imagery to the landing page usually strengthens the FDA question rather than avoiding it, and adds the consent and asset-register obligations described above.

Does organic social have the same restriction?

Organic posts sit outside ad review until they are boosted or used as creative, at which point they are ads. The FDA question about promotional communications does not distinguish between paid and organic in the first place.

How do we prove compliant creative still converts?

Measure at treatment start, not at click or quiz completion, and keep the clinical detail out of the ad platforms. Curve's server-side path, neutral aliases, bridge tokens, and offline conversion uploads exist so that deeper measurement does not require disclosing who is interested in what.

Where to start

Audit every surface for transformation imagery first: ads, landing pages, quiz result screens, email, SMS, and creator content produced on your behalf. For each one found, record where the asset lives, what consent exists, and what claim it is making. That inventory usually settles the internal debate faster than any policy citation.

Then replace rather than delete. Build process-proof, experience testimonial, and honest-logistics creative, and measure it at treatment start so the comparison is fair. Curve gives you server-side collection, per-destination field mapping, hashed identifiers, neutral event aliases, bridge-token attribution, and offline conversion uploads to make that measurement possible without exposing patients. Run the free compliance scanner against your funnel, or visit curvecompliance.com to talk it through.

Reviewed August 2026. This is general information, not legal or regulatory advice. FDA positions and platform policies change; consult qualified counsel about your specific promotional materials.