Google’s 2025 ads safety numbers are enormous on their own. Across the US, the company blocked or removed 1.7 billion ads and suspended 3.3 million advertiser accounts. Globally, those figures hit 8.3 billion ads and 24.9 million suspended accounts. Now the Video Advertising Bureau, a trade group for TV sellers, has given those numbers a new frame: it compares them with the 86,800 brand advertisers and 146 million ad units that ran on national and local broadcast TV in the same period.
The resulting ratio is striking: roughly 38 times more suspended US ad accounts than TV had advertisers. But the comparison is less simple than the headline makes it look.
Where the 38x comparison gets slippery
The VAB’s “Bad Actors” analysis does not include new data collection. It draws on Google’s April 2026 global and US Ads Safety Reports, plus VAB’s Nielsen Ad Intel analysis for the TV side. The arithmetic is accurate, but the units are not equivalent.
- Google counts advertiser accounts, and a single operator can open many; TV counts brand variants.
- Google counts ads that were blocked or removed before they served, while TV counts ad units that actually aired.
- Google Ads is an open self-serve system; TV inventory is sold through negotiated relationships with a limited set of sellers.
Even VAB’s own footnote acknowledges that the biggest policy category includes “creating new accounts to evade enforcement,” which can inflate suspension counts on Google’s side.
The more useful question for marketers
Beyond the TV-vs-digital fight, the deck raises a legitimate transparency issue. Google says its Gemini-powered systems stopped more than 99% of policy-violating ads before they were served. VAB’s response is that “1% can be millions of ads.” On Google’s own numbers, less than 1% of 8.3 billion blocked or removed ads could still mean up to roughly 83 million ads that served before removal. Neither report says how long those ads stayed live, how many people saw them, or where they appeared.
The difference matters for performance teams. A pre-serving review is a prediction about what will run, not proof of what actually ran. Recent malvertising cases, such as Confiant’s SourTrade documentation in July, show why post-bid verification is still necessary. That operation was built to pass review systems while assembling malware in victims’ browsers.
What to do about it
For Google Ads users, the enforcement scale cuts two ways. Google says incorrect advertiser suspensions fell by more than 80% by November 2025, with 99% of appeals resolved within 24 hours. But the operational risks remain. Google Ads ended appeals for policy decisions older than six months in July, and its limited ad serving policy will remove unlimited impressions from unqualified advertisers by 2028. That makes account health, appeal timing, and documentation more important than ever.
Brand suitability also stays on the buyer. The VAB deck shows that 85% of the 480 million pages actioned were restricted for sexual content, a category where ads may still run in limited scenarios. So Google’s enforcement tally is not the same as a brand-safe guarantee.
The takeaway is practical: treat platform safety reports as enforcement metrics, not as proof your campaign environment is clean. Audit account health before appeal windows close, use placement exclusions and post-bid verification for sensitive campaigns, and monitor account-level policy changes closely.
Source: PPC Land
