Google suspended 24.9 million advertiser accounts in 2025, and the categories matter more than the ratio

The Video Advertising Bureau circulated a 13-page analysis on 15 September 2026 that sets Google's 2025 enforcement numbers against linear television, and PPC Land reported it the same day along with a detailed account of where the comparison breaks down. The figures at the centre of it are Google's own: more than 8.3 billion ads blocked or removed worldwide in 2025, more than 24.9 million advertiser accounts suspended, and 3.3 million of those suspensions in the United States alone.

For an advertiser, the useful content here is not the trade body's argument. It is the category breakdown underneath it, which says something concrete about where account suspensions actually come from, and the specific gaps PPC Land identified in how the numbers were presented.

What the document is

The deck is titled "Bad Actors: Examining How Much Objectionable Advertising Google Takes Action Against" and belongs to the trade body's Fast Facts series. It was distributed to media in a weekly newsletter on 15 September 2026 and carries no separate publication date on its pages. Every slide bears a notice restricting the material to members and qualified marketers, and the landing page offers it as both a PowerPoint and a PDF.

No new data collection sits behind it. According to the Video Advertising Bureau, the analysis draws on two Google documents released in April 2026, the 2025 Global Ads Safety Report and the 2025 Ads Safety Report for the United States. The only original figures come from an analysis of Nielsen Ad Intel covering 1 January to 31 December 2025, which supplies the television side. PPC Land, which covered the underlying Google report in April, described the additive element as the frame placed around the numbers rather than the numbers themselves.

The global enforcement figures

Publisher actions are defined in the deck's footnote as ad demonetization, ad termination and account suspension. Scams account for a defined slice of the total: 602 million ads removed for policy violations most closely associated with scams, and more than 4 million accounts suspended for scam-related activity. That is roughly 7% of blocked or removed ads but about 16% of suspended accounts, which suggests scam enforcement leans more heavily on account-level action than the enforcement programme does overall.

Where the removals came from

The category breakdown is the part worth reading if you run ads for a living, because it says which policies actually generate volume.

Abusing the ad network, the largest category, is defined in Google's own terms as covering malicious or unwanted software, hacked websites, gaining an unfair advantage in the ad auction, disguising ad elements to avoid detection, creating new accounts to evade enforcement, and violating Google Search spam policies. Personalization violations cover ads or targeting that use sensitive user data, personal traits or intimate life situations.

The remaining categories, in descending order, are dating and companionship at 354.2 million, financial services at 327.8 million, sexual content at 321 million, gambling and games at 270.7 million, copyright at 229.4 million, inappropriate content at 147 million, healthcare and medicines at 87.8 million, alcohol at 20.6 million, and enabling dishonest behaviour and dangerous products or services at 13.1 million each.

Three gaps PPC Land identified

The first is arithmetic. PPC Land calculated that the 16 listed categories total approximately 5.27 billion ads, about 63% of the 8.3 billion headline, and the deck does not explain the remaining 3 billion or so. Google's policy set includes more categories than the chart displays, and one ad can in principle be actioned under more than one policy, so the gap need not indicate an error. It does mean the chart cannot be read as a complete distribution.

The second is internal. The deck's own Key Marketer Takeaways slide lists trademark and copyright infringement, sexual content, fraudulent actions and counterfeit goods among the top reasons ads were removed. By the deck's own chart, counterfeit goods ranked last of the 16 categories at 513,000 ads and copyright ranked tenth, while personalization violations and legal requirements, the second and third largest categories, do not appear in the takeaway list at all. PPC Land also noted a stray sentence in the PDF's text layer about adults over 50 accounting for 36% of the population, which does not appear on the rendered slide and has nothing to do with the subject.

The third is a missing number. The stated scope covers ads blocked, removed or restricted, but the deck contains no restricted-ad figure. Google's global report put restricted ads at 4.8 billion for 2025, down from 9.1 billion in 2024, according to PPC Land's April coverage. Restricted ads are permitted to run under conditions such as limits on targeting or placement, which is a materially different outcome from removal.

The 480 million pages figure is mostly a restricted category

The publisher-side number invites the reading that nearly half a billion pages were banned from monetisation. The detail says otherwise. Sexual content accounted for more than 409 million pages, about 85% of the total, and the footnote describes it as a restricted category covering partial nudity, sexual merchandise and nudity, where ads may run in limited scenarios.

The rest are far smaller: dangerous and derogatory content at 20.5 million pages, shocking content at 15 million, weapons promotion and sales at 12.8 million, online gambling at 9.7 million, alcohol and tobacco at 5 million each, and sexually explicit content and malware or unwanted software at 2 million each. Those nine categories sum to about 481 million, consistent with the headline. Google said AI-driven systems contributed to detecting more than 467 million of those pages, per PPC Land's April coverage.

For a media buyer, the distinction matters. A page restricted for sexual content can still carry ads under Google's rules, subject to limits, which leaves brand suitability decisions with the buyer's own exclusion controls rather than with Google's enforcement tallies.

The television comparison, and why the arithmetic is not the argument

The slide that gives the deck its headline sets Google's US enforcement against linear television. On the Google side: 1.7 billion ads blocked or removed and 3.3 million advertiser accounts suspended in the US in 2025, representing about 20% of global blocked or removed ads and about 13% of global suspensions. On the television side: 146 million total ad units aired on national and local broadcast TV, and 86,800 brand advertisers, from the Nielsen Ad Intel analysis.

The arithmetic holds. US suspensions are about 38 times the number of TV brand advertisers, and US blocked or removed ads are about 11.6 times the number of TV ad units aired. Whether the comparison measures what the slide headline claims is a different question, and the deck's own footnotes supply the reasons it does not. The television side counts ad units that actually aired, on an unequivalized basis, so a 15-second spot and a 60-second spot each count once. The Google side counts ads that were blocked or removed, most of which, according to Google, were never served at all. The television advertiser count uses Nielsen brand variants, while Google counts accounts.

Counting things that were stopped against things that ran does not produce a like-for-like ratio. A high suspension count is equally consistent with an enforcement system catching a great deal and with an ecosystem attracting a great deal to catch, and nothing in either document separates the two.

PPC Land's own framing of what is still missing is the most useful line in the coverage: Google does not disclose how long the violating ads that slipped through were live, how many people saw them, or where they ran. Those are the numbers that would let anyone judge enforcement effectiveness, and none of them exist in either report.

What this actually means if you run Google Ads

Nothing in Google's rules changed. What changed is that a set of numbers already published in April is circulating again with a sharper frame, so the practical value is in reading the categories rather than the ratio.

• Account suspension is a volume event at Google's scale, not a rare one. More than 24.9 million accounts were suspended globally in 2025, and the largest driver, abusing the ad network, explicitly includes creating new accounts to evade enforcement. A suspended account is not a thing you appeal your way around by opening another.
• Personalization violations were the second largest category at more than 755 million ads, covering targeting that uses sensitive user data, personal traits or intimate life situations. This is the category most likely to catch a legitimate advertiser by accident, particularly in health, finance and dating adjacent verticals.
• Misrepresentation ranked fourth globally at 421.5 million ads and second in the United States by rank. It is the category most directly associated with scams, and also the one that catches overstated claims in ordinary ad copy.
• Healthcare and medicines accounted for 87.8 million ads. Clinics and pharmacies running healthcare marketing operate under a policy set that generates enforcement volume in the tens of millions annually.
• Counterfeit goods, despite being named in the deck's takeaways, was the smallest category at 513,000 ads. Retailers should weight their compliance attention accordingly rather than by what a headline emphasises.
• Brand suitability on the publisher side remains your job. The bulk of the 480 million actioned pages sits in a restricted category where ads may still run, so your own placement exclusions are doing the work you might assume Google's enforcement has already done.

What this means for Thai advertisers

Neither report published a Thailand breakdown, so any figure specific to this market would be invented. What transfers is the structure of the enforcement, which is global.

Two categories deserve local attention. Personalization violations are easy to trip in Thai-language copy that refers to a health condition, a financial situation or a personal characteristic in a way that would read as ordinary marketing locally and as sensitive-category targeting to a policy system. And misrepresentation covers claim inflation, which is common in Thai promotional copy where superlatives are conventional rather than literal. Neither of those is bad intent, and neither is a defence.

The account-evasion point matters more here than it might elsewhere. Opening a fresh account after a suspension is a widespread response among smaller Thai advertisers, and Google's own definition of its largest enforcement category names that behaviour explicitly. For e-commerce sellers in particular, the safer path is fixing the underlying disapproval on the existing account.

Frequently asked questions

How many Google Ads accounts were suspended in 2025?

More than 24.9 million advertiser accounts worldwide, of which 3.3 million were in the United States, according to Google's 2025 ads safety reports as presented by the Video Advertising Bureau. More than 4 million of the global suspensions were for scam-related activity.

What is the biggest reason Google removes ads?

Abusing the ad network, at more than 1.29 billion ads in 2025. Google's definition of that category covers malicious or unwanted software, hacked websites, gaining an unfair advantage in the ad auction, disguising ad elements to avoid detection, creating new accounts to evade enforcement, and violating Google Search spam policies.

Is this new data from Google?

No. The analysis draws on two Google documents released in April 2026, the 2025 Global Ads Safety Report and the 2025 Ads Safety Report for the United States. The only original figures in the deck come from a Nielsen Ad Intel analysis covering calendar year 2025, which supplies the television comparison.

Does the 480 million blocked pages figure mean those sites lost all ads?

No. More than 409 million of those pages, about 85%, fall under sexual content, which the footnote describes as a restricted category covering partial nudity, sexual merchandise and nudity where ads may run in limited scenarios. Restriction and removal are different outcomes.

Do these figures say anything about Thailand?

No. Neither Google report cited in the deck was broken out for Thailand, and the deck's only country-level detail is for the United States. The policy categories themselves apply globally, which is what makes the breakdown useful here even without a local figure.

The short version

A television trade body took Google's own enforcement numbers and built a ratio out of them, and the ratio is arithmetically correct and methodologically mismatched: it counts ads that were stopped against ads that aired. Strip the framing away and the underlying breakdown is still worth an hour of any advertiser's time, because it shows that account suspensions run into the tens of millions, that the largest single category explicitly includes opening new accounts to evade enforcement, and that personalization and misrepresentation are far larger risks for a legitimate advertiser than the categories the deck chose to headline. Advertisers who want their account audited against the policy categories that actually generate enforcement can talk to our Google Ads team.