A disapproved ad reads like a verdict. It usually isn't. The takeaway up front: a disapproval is a review decision about one specific asset, measured against one specific written rule — and every major platform publishes that rule library openly. The productive response is to find the rule the notice names, change the thing it points at, and resubmit the corrected ad for review. The unproductive response is to guess.
This is an evergreen problem rather than a news event: the individual policies change constantly, but the shape of a disapproval and the process for resolving one have been stable for years.
How ad review actually works
Per the published documentation of the major ad platforms, review is layered. Ads are first checked automatically against the platform's policy set, which is why a disapproval can arrive within minutes of publishing. Some ads then get human review, typically when the automated check is uncertain or the category is sensitive. Both stages measure the same thing: the ad, its assets, and where it points.
Two details from that process are worth internalising.
Review covers the destination, not just the creative. Google, Meta, TikTok and Microsoft all publish landing-page and destination requirements alongside their creative rules. An ad can be flawless and still fail because of the page behind it.
Policies are versioned and public. Each platform maintains a policy centre or help-centre library, and each publishes changes to it. If you cannot point at the specific policy your ad was measured against, you are not ready to fix it — that is the first step, not an optional one.
The six causes behind most disapprovals
Across platforms, the same handful of issues account for the large majority of what practitioners see.
1. Claims the ad cannot support. Unverifiable superlatives, guaranteed outcomes, promised earnings, and before/after health or financial results. The rule underneath is consistency: the claim has to be substantiable and it has to match what the destination actually delivers. Softening the wording is not the fix if the underlying claim is still one you cannot evidence.
2. Landing-page quality and mismatch. The destination must deliver what the ad promised, load properly, and behave like a legitimate business page. Common failures: an ad for a specific product landing on a generic homepage, a page missing basic business identity or contact information, absent privacy or refund terms where they are expected, aggressive interstitials, or a destination that does not work on mobile. This is the most frequently misdiagnosed cause, because people rewrite the ad repeatedly when the page was the problem.
3. Categories that require certification or verification. Financial services, healthcare and pharmaceuticals, alcohol, gambling and several others are gated: allowed only with the right local licence, platform certification, or advertiser verification, and often only in specific countries. Nothing you change in the creative fixes this. The path is the platform's certification or verification process, and it takes time you should plan for before the campaign is built.
4. Trademark and brand-name use. Using another company's brand in ad text or as a keyword is governed by separate trademark policies with their own complaint and authorisation processes. Reseller and informational uses are treated differently by different platforms — check the specific policy rather than assuming.
5. Technical and formatting rules. Excessive capitalisation, repeated punctuation, gimmicky symbols, unsupported characters, broken or redirecting destinations, and required disclosures missing from the creative. These are the cheapest to prevent and the most annoying to hit, because they have nothing to do with your offer.
6. Advertiser identity verification. The major platforms have moved steadily toward verifying who is paying for advertising, and unverified advertisers face limits on what they can run. This is administrative rather than editorial, and it is worth completing before you need it.
Fix the asset, not the wrapper
The workflow that resolves disapprovals quickly is unglamorous and consistent.
Read the exact policy named in the notice. Not the summary in the interface — follow through to the policy page. It states the rule and usually gives examples of compliant and non-compliant treatments. Most of the time the answer is there.
Decide which asset is at fault. Headline, description, image, video, extension, or destination. If it is the destination, editing the ad will not help; you need a page change or a different page.
Change the substance. If the claim was unsupportable, the fix is a claim you can support — not a synonym. If the page did not match the ad, either the page changes or the ad's promise does. Resubmitting an asset that has not meaningfully changed wastes review cycles and teaches you nothing.
Then resubmit the corrected ad for review, and record the outcome. Which policy, which asset, what you changed, what happened. After a dozen entries you will see your own recurring pattern, and that pattern is what your checklist should target.
Treat systematic disapprovals as a briefing problem. If the same issue keeps appearing across a campaign, the fault is upstream in how creative is briefed, not in the individual ad.
Build the checks in before you publish
Most disapprovals are preventable at the drafting stage. A short pre-flight list, run before anything goes live:
- [ ] Every claim in the ad is one you could evidence if asked.
- [ ] The destination delivers exactly what the ad promises — same offer, same product, same price.
- [ ] The landing page loads fast, works on mobile, and shows real business identity and contact details.
- [ ] Privacy, terms and refund information are present where the category expects them.
- [ ] No excessive capitalisation, repeated punctuation, or unsupported characters.
- [ ] No third-party brand names used without checking the relevant trademark policy.
- [ ] If the category is gated, certification or verification is already complete — not in progress.
- [ ] Advertiser identity verification is done.
- [ ] Someone other than the writer has read the ad against the policy library.
Policy libraries also change, which is a monitoring problem rather than a drafting one. The practice of watching official changelogs and translating each change into an action is covered in our guide to responding to ad-platform changes — the same triage applies here: confirm the change at source, decide whether it touches your accounts, then act.
Where account structure and support fit
Structure does not change the rules — the same policies apply to an advertiser in a partner or agency structure exactly as they do to a self-serve one. What structure can change is how quickly you get an informed answer to a setup, verification or billing question, and whether a category problem is spotted before you have built a campaign around it.
That is the honest case for a partner arrangement with real support attached. ADShift, which supplies agency ad accounts across Google, Meta, TikTok and Bing, is a reasonable example to look at for the sequence of its onboarding rather than the marketing around it: a business review comes first, before approval, payment and account setup, with an account manager and a support-ticket channel afterwards. A provider that examines what you sell before granting access is applying the same category check you should be applying yourself, and finding out at that stage that your vertical needs certification is far cheaper than finding out later.
Evaluate it as you would any vendor in that position — what the fee covers (its published rates start from 3% on Google, 0% on Meta, 3% on TikTok and 7% on Bing), what support is actually included, and what happens to your accounts and data if you leave. And keep the compliance work in-house regardless: nobody else is going to read your landing page for you.
FAQ
Does a disapproval mean I did something wrong?
Not necessarily. Review is largely automated, applies a broad rule set to an enormous volume of ads, and errs toward caution — so a compliant ad can be caught by a rule it resembles. The correct reading is neutral: a specific asset was measured against a specific policy and did not clear it. Find the policy, check honestly whether the ad meets it, and fix or clarify accordingly.
Are the policies the same across Google, Meta, TikTok and Microsoft?
They rhyme but they differ, and the differences matter. All four prohibit the same broad categories and all four publish landing-page requirements, but the details — what counts as a gated category, which certifications are recognised, how trademark use is handled, which countries a category is allowed in — vary by platform and by country. Check each platform's own policy centre; do not port an assumption from one to another.
What is the fastest way to resolve a disapproval?
Read the named policy first, identify the exact asset at fault, change the substance rather than the phrasing, and resubmit once. The slowest path is repeatedly resubmitting near-identical versions hoping for a different outcome — it burns review cycles and produces no information about what the actual problem was.
How much of this is really about the landing page?
More than most advertisers expect. Destination requirements sit in every major platform's published policies, and a page that is slow, mismatched, missing business identity, or thin on the information its category requires can fail review on its own. When an ad is disapproved and the creative looks clean, check the page before you rewrite the headline again.
Do certification and verification take long?
Long enough that they belong at the start of a campaign plan rather than the middle. Gated categories generally need a valid local licence plus platform certification, and advertiser identity verification is its own process. Neither is something you can shortcut with better copy, so build the lead time into the launch schedule.
The short version
Disapprovals are a normal part of running paid media, not a signal that something has gone badly wrong. Handle them as process, not panic: identify the exact policy, decide which asset it applies to, change that asset in substance, resubmit once, and log the outcome so the pattern becomes a checklist. Put the checks in before publication and most of the problem disappears. If you also want a partner structure with consolidated invoiced billing and a named support contact behind it, ADShift is one provider worth evaluating — on its fees, its support and its onboarding checks, with your own compliance review firmly unchanged.
