That Viral "$442K at 2.5x ROAS" Discover Ad Screenshot Breaks Two Google Ads Policies
A widely-shared screenshot brags about a Discover ad using a before/after hair-loss collage. That image format is explicitly banned for health and cosmetic categories. Here's the actual policy, why the screenshot is misleading either way, and what to run instead.
A screenshot has been making the rounds: a native Discover ad campaign, $442K in revenue, 2.5x ROAS, $0.93 average CPC. The caption calls it one of the poster's top cold-traffic campaigns of the year. The ad creative in the screenshot is a side-by-side before/after photo of a woman's thinning hairline, next to the headline "The real reason your hair looks thinner every morning. Discover how your environment may be affecting growth, and the serum people are switching to."
If you run ecommerce ads and your instinct on seeing that was "I should try this," stop. That image is a textbook violation of two separate Google Ads asset policies, and the screenshot itself doesn't tell you what you think it tells you.
The Two Policies This Ad Breaks
Google's advertising policies on image extensions and asset requirements (support.google.com/adspolicy/answer/10347108, support.google.com/google-ads/answer/9823397) are specific about two things that this creative does at once.
| Policy | What it says | What the ad does |
|---|---|---|
| No image collages | Composite images stitched from multiple photos into one asset are banned, outside naturally occurring photo grids. | Two separate hairline photos spliced side by side into a single image asset. |
| No before/after for health, cosmetic, or weight-loss claims | Comparison imagery for hair loss, weight loss, cosmetic procedures, or beauty/appearance treatments is restricted as an unsubstantiated-claims and clickbait risk. | The before/after framing exists specifically to imply the product reversed hair thinning, a health and appearance claim. |
Either one on its own is enough to trigger an automated disapproval on review. Stacked together, on a category (hair loss serum) that Google's ad-review systems already treat as higher-risk, this creative is not a gray area. It's a clear violation sitting in a Discover feed.
Why the Screenshot Is Misleading Even If the Numbers Are Real
Assume the $442K and the 2.5x ROAS are genuine and not inflated. That still doesn't make this a strategy worth copying, for one reason: automated image review doesn't catch every violation on the first pass, and enforcement on Demand Gen and Discover placements is inconsistent. An ad can run, spend real budget, and generate real revenue for days or weeks before a manual review, a competitor report, or a policy sweep catches it.
A screenshot taken mid-run is proof the ad was live at that moment. It is not proof of a stable, repeatable system, because the account running it is sitting one review cycle away from a creative disapproval at best, and an account-level policy strike at worst. Bragging about the revenue from an ad that's actively violating asset policy is like posting your speed on a highway camera that hasn't mailed the ticket yet.
The Real Risk Isn't the Disapproval
A single disapproved ad just gets pulled. The bigger risk is what repeat violations do to the account's policy standing. Google tracks violation history at the account level, and health/cosmetic misrepresentation is one of the categories that escalates fastest from "ad disapproved" to "account under review" to a Merchant Center or Ads account suspension.
The General Image Rules Behind This
See this on your own account
Two ways Shopiator can help: run the $49 audit yourself, or talk to the founder directly.
Beyond the health/cosmetic before-after restriction specifically, Google's image asset specs are worth knowing cold if you run Shopping, PMax, Demand Gen, or image extensions:
- ▶No collages or stitched panels, single-subject photos only, aside from genuine, unmanufactured photo grids
- ▶No text, graphic overlays, or logos burned into the image itself, that belongs in the ad copy, not the asset
- ▶JPG, PNG, or static GIF only
- ▶Square (1:1) at 1200x1200 recommended, landscape (1.91:1) at 1200x628 recommended
- ▶5,120 KB maximum file size
What to Run Instead
The instinct behind that ad isn't wrong, curiosity-gap headlines and results-driven creative work. The execution is what's illegal under policy. If you're selling anything results-driven, hair growth, skincare, supplements, fitness, the transformation story needs to live in your copy and your landing page, not in a spliced photo.
- 1.Use a single clean product or lifestyle photo as the image asset, not a comparison
- 2.Put the curiosity hook in the headline and description text, where Google's health-claims review is about substantiation, not image format
- 3.Back any results claim with a real substantiation page (study citation, dermatologist review, clinical data) linked from the ad, since unsubstantiated claims get flagged in copy too, just less aggressively than in imagery
- 4.If you want a visual before/after, put it on the landing page behind the click, not in the ad asset Google reviews pre-click
None of this is a workaround. It's just where the line actually sits. The image asset is the highest-scrutiny surface in the whole funnel because it's the first thing reviewed and the easiest thing to automate detection against. Everything downstream of the click has more room.
The Takeaway for Ecom Founders
Viral ad screenshots optimize for engagement, not for being a safe playbook. Before you screenshot someone else's "what's working for us" post into your own account, ask one question: would this creative survive a manual policy review today. If the answer is no, the revenue number in the screenshot is a countdown timer, not a strategy.
We'll flag policy-risk creative, account health issues, and wasted spend before Google does it for you.
Get a Free Google Ads Account AuditFrequently Asked Questions
Are before-and-after images banned in all Google Ads campaigns?
Not universally, but they're heavily restricted, and for weight loss, health treatments, cosmetic procedures, and beauty/appearance claims they're treated as unsubstantiated claims and clickbait bait. If your product touches any of those categories, assume before/after imagery gets the ad disapproved and build creative around it instead of hoping it slips through.
Why would a policy-violating ad still show real spend and revenue numbers?
Automated image review doesn't catch everything on the first pass, and enforcement is inconsistent, especially on Demand Gen and Discover placements. An ad can run and generate real revenue for days or weeks before a manual review or a policy sweep catches it. A screenshot taken mid-run proves the ad was live at that moment. It doesn't prove the campaign is a stable, repeatable playbook, because that same account is one review cycle away from a disapproval or a suspension.
What counts as an image collage under Google's policy?
Any composite image stitched from multiple separate photos into one asset, side-by-side panels, split-screen comparisons, or grids, unless it's a naturally occurring photo grid (like a genuine Instagram-style layout, not manufactured to imply a before/after narrative). A single photo with a caption underneath it is fine. Two photos glued together inside one image file is what gets flagged.
What should I use instead of a before/after image for a results-driven product?
Put the transformation claim in the ad copy or headline, backed by a substantiation page it links to, and use a single clean product or lifestyle photo as the image asset. The curiosity gap that makes before/after ads work can live in the words. It doesn't need to live in a spliced photo that gets the ad pulled.
Ready to fix this on your account?
Two ways Shopiator can help: run the $49 audit yourself, or talk to the founder directly.
Keep reading
The Google Ads Domain-Switch CPC Hack, Explained
A viral thread claims switching your domain drops Google Ads CPCs by 75%. The drop is real, the cause is mundane: smart bidding's confidence reset. Here's the mechanism, why it reverses, what the screenshots actually show, and a 6-point smell test for the next viral 'hack'.
ReadGoogle AdsWhy Most Ecom Brands Fail at Google Ads
Most ecommerce brands waste 15-30% of their Google Ads budget without knowing it. Here are the 7 most common mistakes and how to fix each one.
ReadRanking8 Best Incrementality Testing Tools for Ecommerce (2026)
Blended ROAS lies. Incrementality tells you what your ads actually caused. An independent comparison of the geo-holdout and incrementality testing platforms ecommerce brands actually use, plus how to run a holdout test with tools you already have.
Read