Weight Loss Ad Disapprovals: A Triage Checklist

Most weight loss ad disapprovals trace back to one of six causes: a branded drug term, before-and-after imagery, personal attributes copy, a missing prescription drug authorization, a landing page that fails review even though the ad text passed, or an account-level restriction that no creative change will lift. Triage means identifying which one you have before rewriting anything. Curve is the HIPAA-compliant tracking layer weight loss clinics use to measure these campaigns without exposing patient data, with a signed Business Associate Agreement on every plan, and accurate measurement is what tells you which fix actually worked.

Start by reading the rejection properly

The stated reason is a category, not a diagnosis. Meta and Google both return short policy labels that map to broad rule families, and the same label can cover several distinct problems. Before you touch the creative, collect four things.

1. The exact policy label as shown in the platform interface, not your paraphrase of it.
2. Which asset was rejected. The ad, the ad set, the whole campaign, or the account. This distinction determines whether a creative fix is even possible.
3. What changed immediately before. New creative, a new landing page, a new audience, a new payment method, a new page admin, or nothing at all. Rejections with no local change usually mean a policy update or an account review.
4. Whether similar ads are still running. If a near-identical ad is live, the difference between the two is your evidence.

Then work the list below. It is ordered by frequency, so the first match is usually the real one.

Cause 1: a branded pharmaceutical term

Signals. Rejection cites prescription drugs, unapproved pharmaceuticals, or restricted content. The ad, the headline, the image text, or the destination page names a brand-name weight loss drug.

What is happening. Meta rejects branded pharmaceutical weight-loss terms. The check runs on the creative and on the landing page, so a clean ad pointing at a page built around a brand name inherits the problem. It also catches near-misses: deliberate misspellings, the brand name inside an image, and phrasing such as "the same as" a named drug, which asserts an equivalence that has not been established.

Fix. Remove the brand name from every surface the platform can see, including image text, the URL path, the page title, and the meta description. Describe the category of care and the medical evaluation instead of the product. If the compound is compounded rather than approved, equivalence framing is also an FDA problem independent of platform policy, which we cover in our piece on compounding pharmacy GLP-1 advertising restrictions.

Cause 2: before-and-after imagery

Signals. Rejection cites personal health, unrealistic outcomes, or adult content. Creative contains a split image, a transformation sequence, a scale, a tape measure, or a close crop of a body part.

What is happening. Meta rejects most before-and-after weight-loss imagery outright, and the detection is broader than the literal split-screen format. A single photo can read as a transformation claim when the copy frames it that way. Zoomed body crops are flagged under a separate rule about depicting body parts to imply an outcome.

Fix. Replace outcome imagery with process imagery. The clinic, the provider, a consultation, a lab result, a product-free lifestyle image. If social proof is essential, use written testimonials with typicality context rather than photos, and keep in mind that outcome claims in text carry substantiation expectations too.

Cause 3: personal attributes copy

Signals. Rejection cites personal attributes or personal health. Copy addresses the reader in the second person about their body, their weight, or their condition.

What is happening. Platforms prohibit copy that asserts or implies knowledge of an individual's health status. "Struggling with your weight?" is borderline. "You have tried everything and nothing works" is an assertion. The rule is not about the sentiment, it is about who the sentence claims to know something about.

Fix. Rewrite from the second person to the third person or to a question about the category. "Medical weight management, supervised by licensed providers" is safe. "Wondering whether medical weight management is right for you" is usually safe. Anything that tells the reader what is true about their body is not.

Cause 4: missing prescription drug authorization

Signals. Every prescription-adjacent variant is rejected regardless of wording, while general wellness copy runs fine. Rejection cites prescription drugs or a certification requirement.

What is happening. Meta requires prior authorization for prescription drug advertising and admits only pharmaceutical manufacturers, online pharmacies, and telehealth providers. Google gates prescription drug promotion behind certification limited to specific entity types and countries. Without the relevant permission, no rewrite unlocks the category, because the limitation is on the advertiser rather than on the ad.

Fix. Determine whether your entity actually qualifies. Many medical weight loss clinics operate a telehealth arm that meets Meta's definition and never applied. If you qualify, apply, and expect the review to examine your clinical model rather than your creative. If you do not qualify, build the funnel around the consultation and the clinical program, which remains fully advertisable. Our GLP-1 advertising policy update for Google and Meta covers where those lines currently sit.

Cause 5: the landing page, not the ad

Signals. The same creative is approved in one ad set and rejected in another with a different destination. Rejection cites low quality destination, misleading claims, or circumventing systems.

What is happening. Both platforms review the destination. Common landing page triggers in this category: a specific weight loss figure without substantiation, a countdown timer or scarcity device, pricing that appears only after a quiz, an interstitial before the real content, missing risk information, unclear provider identity, and a brand drug name anywhere in the page source. A page can also fail because it redirects, because it differs on mobile, or because it requires an email address before showing anything.

Fix. Load the page the way a reviewer will: fresh session, mobile viewport, no ad blocker. Read what is visible before the first scroll. Then remove unsubstantiated numbers, surface pricing and risk information, name the provider and their licensure, and delete artificial urgency. We wrote up the structure that works in GLP-1 clinic landing pages that convert without PHI.

Cause 6: account or page level restriction

Signals. Every ad is rejected instantly, including ones previously approved. New ads fail before review could plausibly have happened. Business manager shows a restriction notice.

What is happening. Accumulated policy violations, a page-level restriction, a billing or identity verification issue, or an association with another restricted asset. Creative changes are irrelevant at this level.

Fix. Go to the account quality view and read the actual restriction. Appeal at the account level, not per ad. Do not spin up a new account to route around a restriction, because evading enforcement is itself a violation and it typically ends with both assets restricted. Fix the underlying pattern first, then appeal once with a coherent explanation.

The triage order, condensed

1. Is the whole account or page restricted? If yes, stop. Nothing creative matters until that is resolved.
2. Does any surface name a branded drug, including image text and the URL? Remove it everywhere.
3. Does the creative show a body outcome? Replace with process imagery.
4. Does the copy tell the reader something about their own body? Rewrite in the third person.
5. Is every prescription variant failing while general copy passes? This is an authorization gap, not a copy problem.
6. Does the same creative pass elsewhere with a different destination? The landing page is the cause.
7. Still unclear? Split the variables. One change per ad, launched together, so the rejection pattern isolates the trigger.

Keep a written log of every rejection with its label, the asset, the change that preceded it, and the outcome of the fix. Within a couple of months that log is worth more than any published policy summary, because it reflects how review actually behaves on your account.

The problem triage cannot see

Fixing disapprovals requires knowing which creative and which landing page variants actually produce patients, so you can retire aggressive versions without destroying the funnel. That requires measurement, and measurement is where weight loss advertisers create a second exposure that no disapproval ever warns them about.

A standard Meta Pixel or Google tag on a weight loss funnel sends the page URL with every event, and those URLs routinely name the program or the medication. It sends quiz completions with answers attached if the form is instrumented naively. It sends a persistent identifier tying the sequence to one person. Meta and Google do not sign BAAs for their advertising products, so 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 this mechanism has cumulatively produced more than $100 million in settlements. An account with a spotless policy record can be carrying the full exposure.

How Curve keeps the measurement clean

Curve is HIPAA-compliant ad tracking, attribution, and analytics built for healthcare. It replaces client-side pixels with a server-side path, which puts a decision point between your site and the 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 instead of straight to the platforms.
• Per-destination field mapping. Only explicitly mapped fields forward. Program names in URLs, quiz answers, and free-text fields stay behind by default.
• Neutral event aliases. The platform sees a neutral conversion name rather than one identifying the weight loss program, which also keeps the service line out of the ad account interface.
• Identifier hashing. Email, phone, and name are SHA-256 hashed per each platform's conversion API requirements.
• Bridge tokens. Attribution survives the handoff to a separate intake or booking tool, which is where these funnels normally lose the chain.
• PHI-pattern detection. Payloads are monitored for PHI-shaped values such as SSNs, MRN-style identifiers, dates, and long numeric sequences, so a changed quiz field surfaces as a flag.
• Offline conversion uploads. Bulk CRM or EHR upload with click ID matching, so optimization targets patients who start a program rather than people who complete a quiz.

Clean conversions forward server-side to Meta CAPI, Google Ads Enhanced Conversions, TikTok Events API, Microsoft, and LinkedIn, with a signed BAA on every plan. For the implementation shape on Meta specifically, see our Meta Conversions API architecture for healthcare.

Frequently asked questions

Should we appeal or edit?

Appeal when you believe the rejection is wrong and you can say why in one sentence. Edit when you can identify a plausible trigger. Repeated appeals on ads that genuinely violate policy contribute to account-level restriction, which is the outcome you most want to avoid.

Why did an ad that ran for months suddenly get rejected?

Policy enforcement is periodically re-run against live inventory, and enforcement thresholds move. A retroactive rejection with no local change usually means the rule tightened rather than that you did something new. Check whether the landing page changed too, since page edits are the most commonly forgotten variable.

Can we use the drug name if we sell a compounded version?

No, on two counts. Meta rejects branded pharmaceutical weight-loss terms regardless of what you dispense, and describing a compounded product with an approved drug's brand name is an equivalence claim the FDA has repeatedly objected to in warning letters to telehealth companies.

Are written testimonials safer than before-and-after photos?

Generally yes for platform review, since the imagery rule is specific. They are not exempt from substantiation and typicality expectations, so a testimonial presenting an unusual result as ordinary carries the same claim risk in text form.

How many variants should we test at once?

Enough to isolate one variable each. Four ads differing in exactly one element tell you more than twenty ads differing in several. The point of the exercise is diagnosis, not volume.

Does fixing disapprovals also fix our compliance risk?

No. Ad review is platform policy. Pixel exposure is a separate legal question governed by HIPAA and state privacy law, and an approved ad can point at a landing page that transmits a patient's treatment interest to a vendor with no BAA. The two need separate remediation.

How do we tell what our pages are currently transmitting?

Open a landing page with the browser network tab active and read the outbound requests to ad platform endpoints, checking the URL parameter, the referrer, and any form values. Our free scanner runs a faster version of the same check.

Where to start

Run the triage order top to bottom on your next rejection instead of rewriting from instinct. Confirm the account is healthy, strip branded terms from every surface, replace outcome imagery, move copy out of the second person, check whether an authorization gap explains the pattern, and load the landing page as a reviewer would. Log every result, because the log becomes your policy map.

Then fix the part no disapproval will ever tell you about. Curve replaces client-side pixels with server-side collection, forwards only explicitly mapped fields, hashes identifiers, uses neutral event aliases so the platform never learns the program, and holds attribution across a separate booking tool, with a signed BAA on every plan. Run the free compliance scanner against your weight loss funnel, or visit curvecompliance.com to walk through it with our team.

Reviewed August 2026. This is general information, not legal advice. Platform policies change frequently; consult qualified counsel about your specific promotional materials and data practices.