Meta rejected your ad. Its own policies say which part of the creative to change.

Meta's own instruction for a rejected ad is more specific than the advice written about it:

We recommend that you review the policy-violating components, like the image, video, text, targeting or landing page and update them to comply with the Advertising Standards.

Components. Most creative rejections come down to one element — a crop, a sentence of overlay copy, a claim, sometimes a targeting setting that isn't in the creative at all — and the fix is to change that element rather than to rebuild the ad. Two of the beliefs most commonly repeated about which elements those are turn out to be wrong when you read Meta's current policies: before-and-after imagery is explicitly permitted, and there is no longer any limit on how much text an ad image can contain.

This post is about the creative only. Ads also get rejected or restricted over landing pages, billing, account history and repeat violations — all real, all documented, none of them fixable by editing an image. Meta's review "may include … an ad's associated landing page or other destinations", and repeat violations restrict the business account rather than the ad.

Before-and-after is on the allow-list

The belief is that Meta bans before-and-after images. Page one still says so: one widely-linked agency post, dated May 2022 and still ranking, lists "No before and after shots (most commonly seen for weight loss)".

Meta's live Health and Wellness standard says the opposite, in its allow list — and says it twice, in two independent documents. From the Transparency Center:

When targeting people 18 years or older, advertisers can run ads that promote: … General cosmetic products, procedures, surgeries depicting before and after transformation.

And from the Business Help Center version of the same policy:

General cosmetic products, procedures or surgeries that depict before and after transformations.

Demonstrating results for weight products is allowed on the same terms: "The ad can demonstrate use of the product or the service, the impact of use or a clear indication of the time it takes to achieve noticeable results."

Outside health and cosmetics the question doesn't even arise. The phrase "before and after" appears nowhere in the Advertising Standards index, nowhere in the Prohibited Commercial Practices standard, and nowhere in the corresponding Community Standard — all checked in full. A before-and-after of a kitchen remodel, a driveway, a carpet or a dashboard has no policy to comply with.

So what does get a before-and-after rejected? Four things, and only one of them is the layout's fault:

• The crop. Ads may not "depict a close up on a specific body area with pinched fat."
• The framing text. No "statements of inferiority about physical appearance" — Meta defines these as "terms, descriptions or questions that are negative and attack an individual's appearance, specific body parts or hygiene." And no "clickbait tactics in a health, weight loss or weight gain context, such as sensational language with exaggerated or extreme claims, or promises of specific outcomes within a set timeframe without disclaimers or qualifiers."
• The claim. Not allowed: that "results can be achieved solely by use of wearable products", or skin whitening products "that cause permanent skin color change".
• The targeting. Health, diet, weight and cosmetic ads "must be targeted to people at least 18 years or older". That is a campaign setting. No amount of editing the image fixes it.

The practical read: change the caption and the crop, keep the layout. The before/after structure is one of the most reliable ways to show a product doing its job, and the policy is not aimed at it. It is aimed at the language wrapped around it.

There is no text limit on ad images

Meta's Best practices for image ads page opens with this note:

Note: There is no longer a limit on the amount of text that can exist in your ad image. The text overlay tool is no longer available.

Present tense, live page. The help article that every guide still cites for the 20% rule now renders under a different title — it is a safe-zone page — and contains no mention of "20%", a text ratio, or a text rating of any kind.

We also went looking for the "Image Text Rating" scale of OK / Low / Medium / High that page one repeats as a live throttling mechanism, and could not find it on any current Meta page. That is a statement about what we could locate, not proof that nothing like it exists internally — but if you are being told your reach is capped by a text ratio, nobody is citing a Meta page that says so.

What Meta does still publish about text on images is qualitative: "Use a modern, clean font in a large enough type size and a contrasting hue", don't obstruct the visuals, "don't communicate too many messages", and stay inside the safe zone. The safe zone is the rule that replaced the text rule in practice, and it is stated in percentages rather than pixels. For Reels ads carrying disclaimers, Meta asks for the bottom 40% to be left clear.

The one delivery consequence Meta does document is much vaguer, and worth knowing because it is not a rejection: "Lower quality ads which do not necessarily violate our policies may experience impacted performance."

(The 20% rule is one of several claims about ad creative that outlived their source. We traced eight more of them.)

The element map

Meta's creative policies, and the part of the ad each one is actually about:

Two of those deserve expanding, because Meta publishes unusually concrete guidance on them that almost nobody quotes.

Personal attributes: Meta publishes the rewrites

This is the policy that catches ordinary, well-meant copy — anything that implies you know something about the person seeing the ad. Meta's own help page gives violating examples with suggested alternatives:

Example ad text: "Meet other Jewish singles!" — Violating content: "This ad implies that the individual's religion is Jewish." — Alternative ideas: "Take this as an opportunity to highlight what someone who joins this group can expect as a member."

Example ad text: "Find out how our financial services can help people like you who just turned 62…" — Violating content: "This ad implies knowledge of the exact age of the person who sees the ad." — Alternative ideas: "Explain how your financial services are positioned to help potential customers, including seniors."

The pattern is consistent: move the sentence from who you are to what this does. And one clarification that saves a lot of pointless rewriting — the word "you" is fine:

Note that you can use the words "you" or "your" as long as your ad doesn't mention any prohibited personal attributes.

Meta also notes that asking a question about a personal attribute is the same violation as asserting one. "Struggling with back pain?" implies a medical condition just as surely as stating it.

Sensational content: a four-item checklist

Meta's Sensational Content page is short enough to use as a pre-flight check, and it is about attention tactics rather than gore:

We don't allow shocking, sensational or excessively violent content in ads, including the use of shocking or scary tactics to grab people's attention.

Then the four: shock, scare or disgust; graphic violence or torture; highlight suffering; depict the brandishing of firearms. If a scroll-stopper concept depends on making someone flinch, that is the policy it runs into.

The review and appeal path, and the trap in it

Timings, from Meta:

• Initial review: "Most ads are reviewed within 24 hours, although in some cases it may take longer." It is primarily automated.
• Re-review after an appeal: "We aim to review and provide a decision within 48 hours", and "Unlike the initial ad review, we rely more heavily on teams of human reviewers to process re-review requests."

Then the parts people learn the hard way.

You get one appeal per ad. "If it's rejected again, it will stay rejected and you won't be able to submit another review request for the same ad."

Deleting the ad forfeits the appeal. "If you delete the ad, you may lose your ability to request another review if you believe your ad was incorrectly rejected." The instinct to clear a rejected ad out of the account is the wrong one if you think the rejection was a mistake.

Editing doesn't clear the violation. "Editing or deleting an ad that was rejected for violating Advertising Standards will not remove the violation from your account and you can be restricted from advertising for severe or repeat ad violations." An edited ad is treated as a new ad and reviewed again — the original violation stays on the record either way.

Appealing costs you nothing. "Requesting another review of an ad will not hurt your standing with Meta in any way." Worth stating, because the fear that it does is common.

One genuine inconsistency to be aware of: Meta's two documents disagree about where you appeal. The Transparency Center's Advertising Standards say you "can request a review of the decision in Account Quality". The Business Help Center's troubleshooting article says to request another review in Business Support Home, with click-by-click steps. Page-one blogs say a third thing, Ads Manager, which matches neither. We found no Meta source saying one destination replaced the other. The Help Center article is the one with the actual steps, so that is the safer place to start.

The order to check things in

1. Read which component was flagged. The rejection notice names it, and it is the difference between a five-minute fix and a rebuild.
2. Check the copy before the image. Personal attributes, inferiority statements and timeframe promises live in text, and text is the cheapest thing to change.
3. Check the crop. Body-area close-ups are a named prohibition; the rest of the composition usually isn't.
4. Check whether it's the creative at all. Age targeting on a health or cosmetics ad is a campaign setting. So is the landing page.
5. Appeal before you edit, if you think it's wrong — one shot per ad, and deleting can forfeit it.
6. Don't rebuild around rules that no longer exist. No text limit, no before-and-after ban.

Sources

All pages read in a rendered browser on 20 September 2026. Meta's Business Help Center articles display no publication or revision date; the Transparency Center policy pages show a change-log widget whose individual entries we could not open.

• Meta Advertising Standards — Transparency Center. The policy families, the review process, "the specific components of an ad, such as images, video, text and targeting information", "Lower quality ads … may experience impacted performance", the negative-self-perception principle, the Disruptive Content rule for video, and the Account Quality appeal instruction (accessed 2026-09-20)
• Health and Wellness advertising standard — Transparency Center. The before-and-after allow-list entry, the pinched-fat and inferiority prohibitions, the wearable-results and clickbait-timeframe rules, the 18+ targeting requirement (accessed 2026-09-20)
• About Meta's Health and Wellness advertising policy — Business Help Center. The second, near-identical statement that before-and-after transformations may be advertised (accessed 2026-09-20)
• How to troubleshoot a rejected ad — Business Help Center. "Review the policy-violating components"; the 48-hour re-review; one review request per ad; deleting forfeits the appeal; editing does not remove the violation; appealing does not hurt your standing (accessed 2026-09-20)
• About ads in review — Business Help Center. "Most ads are reviewed within 24 hours"; which edits trigger a new review; "both machines and people make mistakes" (accessed 2026-09-20)
• Personal Attributes — Business Help Center. The worked violating examples with suggested rewrites, and the "you"/"your" clarification (accessed 2026-09-20)
• Sensational Content — Business Help Center. The four-item prohibition list (accessed 2026-09-20)
• Best practices for image ads on Facebook and Instagram — Business Help Center. "There is no longer a limit on the amount of text that can exist in your ad image"; the qualitative text-overlay guidance (accessed 2026-09-20)
• About text overlays and the safe zone for ads on Facebook and Instagram — Business Help Center. The page that historically carried the 20% rule and now carries none; the "leave the bottom 40% free" guidance for Reels ads with disclaimers (accessed 2026-09-20)

---

Every policy quoted above was read on Meta's own site; Meta's help pages are client-rendered, so they were read in a real browser rather than fetched. The searches behind the negative findings — no general before-and-after prohibition, no text-density limit, no Image Text Rating scale — were run over the full text of the Advertising Standards index and the relevant policy pages, plus Meta's own help-centre search. They establish what we could not locate on Meta's current documentation, which is not the same as proof that no such rule has ever existed or is applied internally. None of this is legal advice, and Meta revises these pages without notice.

Terrads composes each placement for its own frame, with the copy and the crop under your control rather than baked into a template.