Can You Put a Before and After Photo in a Paid Ad? (Meta, Google and TikTok, Compared)

You have been posting the same kind of pair for months. A driveway, a client’s hair, a room, nothing dramatic, just an honest before and an honest after. Nobody has ever flagged it. Then you click Boost this post, or you set up a proper campaign for the first time, and it comes back rejected with a policy reason that does not explain much.

What changed is not your photo. It is which rulebook it is being read under. Every platform reviews an organic post under one set of guidelines and a paid ad under a stricter, separate one, and on the question of before and after imagery specifically, the three platforms a trade business actually buys ads on do not agree with each other at all. One of them just rewrote its rule entirely. The other two did not move.

Meta rewrote the test: from what the photo shows to what the ad claims

Until recently, Meta’s health and wellness advertising rules treated before and after imagery as a red flag on sight, regardless of what the ad said. That changed on 22 July 2026, when Meta updated its Health and Wellness Advertising Standards to move from product-based enforcement to claims-based enforcement.

The practical difference is this. Before and after images and side by side comparisons are no longer automatically rejected the moment Meta’s review sees them. What gets judged now is the specific claim the ad is making. A clean, honest before and after of a real result, with no promise attached to it, can run. The exact same image sitting next to lose 30 pounds in 30 days, guaranteed still gets rejected, because the claim is the problem, not the photograph.

Some categories stay off limits regardless of the claim: pinched-fat imagery, sensational health claims, and anything implying a cure for a listed incurable or terminal condition remain banned outright under Meta’s standards. So the change is real, but it is not a green light. It replaces one blunt rule with a narrower one that still has teeth, and the teeth are in the copy, the caption and the claim attached to the image, not in the image alone.

Google still says no, full stop, for three categories

Google Ads has not moved. Its advertising policy for healthcare and medicines prohibits before and after comparison photos for weight loss products, cosmetic procedures and health treatments, and the prohibition does not bend for photos that are genuine, unedited and taken from an actual client. Google’s stated reasoning is that this category of image sets an expectation of a typical outcome by its shape alone, independent of whether any specific claim is attached to it.

That is a meaningfully different design decision from Meta’s. Meta asks what the ad says. Google, for these three categories, does not ask: the image itself is the violation. If your trade sits inside weight loss, cosmetic procedures or health treatments and you are planning a Google campaign, a before and after pair is not the asset to build it around, regardless of how well it was shot or how honestly it was captioned.

TikTok bans it outright, with one narrow exception

TikTok’s position is the strictest of the three. Its advertising policies prohibit before and after comparisons across ad categories generally, including AI-altered imagery, and the ban is not limited to health or cosmetic products the way Google’s is. A dramatic transformation cut used as an ad, in almost any trade this site covers, is a rejection.

The one exception TikTok carves out is for sports and fitness equipment ads, which may use a before and after body comparison as long as the surrounding claims are not exaggerated. It is a narrow lane built for a specific product category, not a general permission. Skincare, supplements, salons, cosmetic services and every trade on this site outside that one exception sit outside it, and TikTok’s separate weight management policy treats a before and after image implying a product caused weight loss or muscle gain as a high-risk claim in its own right.

What boosting a post actually does

Every trade guide on this site assumes an organic post: you shoot a pair, you post it, people who follow you or land on it through a hashtag see it. That is reviewed under a platform’s ordinary content guidelines, which are considerably more permissive than its advertising policy.

The moment you pay to put that same post in front of people who did not choose to see it, whether that is a proper ad campaign or the Boost this post button on a page you already run, the platform reviews it as an ad. That is not a technicality. It is the entire reason a pair that has sat quietly on your feed for a year can be rejected the first time a card is attached to it. The image did not change. The rulebook did.

If your trade is one of the categories named above under Google or TikTok, that gap matters enormously, because it means the safest organic content on your feed can be the least usable content in your ad account.

The register that survives review

None of this means paid before and after content is closed to every trade on this site. It means the register has to be deliberate rather than accidental.

For trades outside the restricted categories, such as pressure washing, landscaping, auto detailing or renovation, the practical risk is lower on all three platforms, and Meta’s claims-based shift makes an honest, unhyped pair the safest version of the format to run. Keep the caption descriptive rather than promissory: what was done, not what it will do for the viewer.

For trades that sit closer to the regulated categories, such as a dental practice, a clinic, or a coach whose content edges toward a body claim, the safer paid asset is usually not a before and after pair at all. A studio fit-out, a piece of equipment, a process shot, or a genuinely non-claim frame carries the marketing job without carrying the risk. The professional template category on this site, including the clinical and dental templates, was built for the register that reads as a record rather than a promise: fixed distance, neutral background, nothing flattering in one frame and not the other. That register happens to be the one every platform’s policy is least suspicious of, for the same reason it looks like a record and not an advertisement.

The underlying discipline is the same one the rest of this site argues for organic posts. A pair that looks manipulated, over-lit in the after, or staged worse in the before, reads as a claim whether you meant it as one or not, and that is exactly the kind of signal an ad reviewer is trained to catch. If you have not already, why your before and afters look fake covers the seven physical tells that make a pair look like a promise instead of a photograph, and fixing them is good practice for organic reach and paid review at the same time. You can check how a pair actually holds together before you spend anything on it with the free before and after slider in your browser.

What this page will not give you

It will not give you a table of what is allowed in which country, and it will not try to predict whether your specific ad gets approved. Ad policy is reviewed by platforms that change it without notice and enforce it through a mix of automated systems and human reviewers who do not always agree with each other, and a business built around predicting the outcome of that review is promising a certainty nobody can actually deliver.

What you can do is check the source directly before you spend money: Meta’s Advertising Standards, Google’s Advertising Policies Help for healthcare and medicines, and TikTok’s Advertising Policies, all published and all updated more often than any third-party summary of them, including this one. If your trade is regulated, a clinic, a dental practice or anything adjacent to a health claim, that check belongs alongside advice from someone qualified in your jurisdiction, not instead of it.

Our app exports a still image, a video or a GIF from two photographs you already have, and it has no opinion on what a platform will approve. What it can do is help you shoot and finish the honest version of the pair, the one built to survive scrutiny rather than invite it. You can see what it does on the app page .