Google August 2026 Spam Update: What B2B Sites Should Do This Week

Google released the August 2026 spam update on August 18. The Search Status Dashboard logged it as an incident affecting Ranking, beginning at 09:27 US/Pacific. At 09:28 PDT, Google posted the only official note: the update applies globally and to all languages, and the rollout may take a few days to complete.

That is the entire primary source. There is no matching Search Central blog post. Google did not announce new spam policy types. And Google has not said this update specifically targets AEO, GEO, or attempts to manipulate AI Overviews and AI Mode.

If your traffic moved on or after August 18, start in Search Console. If it did not, do not panic-publish a 40-page fan-out library to get ahead of it. That behavior is already against policy.

What did Google actually release on August 18?

Google released a global spam update. The dashboard is the source of record. The incident began at 2026-08-18 09:27 US/Pacific. The 09:28 PDT note is one sentence: "Released the August 2026 spam update, which applies globally and to all languages. The rollout may take a few days to complete."

Search Engine Journal and Search Engine Land both confirmed the same dashboard note. SEJ called it the third spam update of 2026. SEL noted the last one was June 2026. SEJ also noted Google had not posted a matching blog article, and that the spam updates page had not changed since December.

Spam updates are not core updates. They are meant to demote or remove content that violates Google's spam policies. They can move rankings while they roll out. A single day's data is not the full picture.

Google will update the dashboard when the rollout is complete, the same way it closed the March and June 2026 updates. SEJ's recap of those two: March took 19 hours and 30 minutes. June took 2 days and 1 hour. "A few days" is the official range. Treat anything before the dashboard closes as incomplete.

Does this spam update target AEO or GEO tactics?

No. Google has not said the August 2026 spam update targets AEO or GEO tactics.

That is the line that matters. Some operators hoped it would, because May 15 already made spam policies apply to generative AI responses. Hope is not a Google statement. The August note does not name AI Overviews, AI Mode, query fan-out, or inauthentic mentions.

What Google did say, earlier this year, is still in force:

• On May 15, 2026, Search Central updated its docs: "Clarifying that spam policies apply to generative AI responses in Google Search." That covers AI Overviews and AI Mode. Source: Google Search documentation updates.
• The spam policies page now defines spam as techniques that deceive users or manipulate Search systems, "such as attempting to manipulate Search systems into ranking content highly or attempting to manipulate generative AI responses in Google Search."
• The AI optimization guide is explicit: creating a separate page for every fan-out query primarily to manipulate rankings or generative AI responses is scaled content abuse. Seeking inauthentic mentions is called out as a tactic you can ignore, because spam systems already block it.

SEJ was careful on this point: the June 2026 spam update followed the May 15 clarification, and Google has not said whether August specifically targets that behavior. Same rule applies here. Do not write "Google just hit GEO spam" unless Google says that.

AEO still matters. The policy already applies to the surfaces your buyers read. The update does not have to name AEO for scaled AI-answer farms and fake mention campaigns to be in scope.

What already counts as spam in AI Overviews and AI Mode?

Google already treats attempts to manipulate generative AI responses in Search as spam. That is not new this week. It was clarified in May.

The policy B2B teams trip most often is scaled content abuse: many pages generated for the primary purpose of manipulating rankings, not helping users. The docs name the method: "Using generative AI tools or other similar tools to generate many pages without adding value for users."

That is the same risk as Does Google Punish AI-Written Articles in June 2026. Google does not punish a page for being AI-drafted. It punishes volume without value.

The AI optimization guide adds the AEO-specific version. Query fan-out is real. Google's systems split a question into related sub-searches. Creating a page for every one of those sub-queries, primarily to manipulate rankings or AI responses, is scaled content abuse. It is also a weak strategy. Google says a high quantity of pages does not make a site higher quality, and the models have improved at understanding relevance without an exact-match page.

Inauthentic mentions are the other trap. Google's guide says generative AI features can show what is being said about products across the web, but "seeking inauthentic mentions across the web isn't as helpful as it might seem." Ranking systems focus on high-quality content. Other systems block spam.

If you shipped 40 near-identical "what is X for Y industry" pages last month to win AI Overviews, that is the pile to review. If you have extractable comparison, pricing, and proof pages that a buyer would actually use, that is not the pile.

For how citations actually work on the two Google surfaces, see today's dual-surface post: Google AI Mode vs AI Overviews. URL overlap is low. Watch them separately.

What B2B sites should check this week

Do not wait for a blog post that may not come. Check the site against existing policy, then watch Search Console from August 18.

1. Search Console from August 18. Compare queries, pages, and countries from August 18 forward against the prior two weeks. SEJ's advice is right: a single day can lie. Look for a pattern that lines up with the rollout, not a Tuesday dip.
2. Scaled, thin, and AI-only pages. Export the URLs published in the last 90 days. Anything that is a synonym swap, a city-page clone, or a raw model dump with no first-hand detail is the risk set. Google's scaled content abuse examples include gen-AI pages without added value, scraped or stitched pages, and multiple sites used to hide the scale.
3. Do not panic-publish 40 fan-out pages. If you are not in an AI Overview this morning, a new library of exact-match sub-query pages is the wrong move. That is the behavior the May docs already called out. Refresh the commercial pages you already have. Make the answers extractable. That is the AI Overviews playbook, not a new calendar.
4. Watch AI Overview and AI Mode citations separately. A ranking move and a citation move are not the same event. Overviews and AI Mode already cite different URLs most of the time. Log the prompt, the surface, and the cited URL. Re-run at 14 days. Citations also churn on a weeks-long clock, which is why AI citations disappear after 30 days.
5. Do not confuse "AI-written" with "spam." If a revenue page was drafted with a model and then edited with real numbers, a named ICP, and a point of view, that is not the August story. The June post covers that line.

A free AI SEO audit will tell you whether crawlers can even reach the pages you are about to review. It will not tell you if this update moved you. Search Console will.

If traffic moved after August 18 and you want a second set of eyes on which URLs look scaled versus which ones just need a refresh, book a 30-minute call.

What Austin Heaton would do if traffic moved

If a B2B site I run dropped after August 18, I would not start a 40-post blog sprint. I would do this, in order:

1. Confirm the timing. Search Console, August 18 onward, by page. If the drop is sitewide and started before the 18th, this is not the August spam update.
2. Isolate the URLs. Which pages lost impressions or clicks? Are they thin templates, AI-only hubs, doorway-shaped city pages, or the comparison and pricing URLs that actually convert?
3. Check existing policy, not rumors. Read the spam policies against those URLs. Scaled content abuse, doorway abuse, and site reputation abuse are the three I see most on B2B sites. The August note did not add a fourth.
4. Cut or noindex the junk. Google's own scaled-content guidance: if you are hosting that content, exclude it from Search. Do not "refresh" 80 clone pages. Remove them or fold the useful bits into one page a buyer would read.
5. Hold the commercial pages. Refresh extractable facts on comparison, alternatives, pricing, and proof. Watch Overviews and AI Mode as two logs. That work lives on the SEO and AEO services page if you want it done, not just diagnosed.

Recovery is slow. Google's documentation, as SEJ noted, says recovery can take months because automated systems need time to see that a site follows the rules. Do the cleanup this week. Do not expect a bounce next Tuesday.

Ready to review the URLs that actually moved, instead of guessing from a dashboard screenshot? Book a discovery call with Austin Heaton.

Read Next:

• Google AI Mode vs AI Overviews: How B2B Brands Get Cited in Both
• How to Get Cited in Google AI Overviews: The B2B Playbook for 2026
• Why Your AI Citations Disappear After 30 Days
• Does Google Punish AI-Written Articles in June 2026?

Frequently Asked Questions

Is the rollout done?

No. Google said the rollout may take a few days. The dashboard is still the source of record. March 2026 closed in 19 hours 30 minutes. June 2026 closed in 2 days 1 hour. Wait for the dashboard to mark it complete before you call the movement final.

Will a good site get hit?

A site that is not violating spam policy should not be the target. SEL noted there will always be cases of sites that were not spamming and still moved. If you got hit and the pages are thin, scaled, or AI-only, treat it as a policy problem. If the pages are extractable commercial URLs with real proof, wait for the rollout to finish before you rip them up.

How long to recover?

Google's documentation says recovery can take months. Automated systems need to recrawl and see that the site now follows the rules. Cleanup this week. Do not expect a same-week rebound.

Does this change how to get cited in AI Overviews?

No. Google did not announce a new citation rule. The May 15 clarification already applied spam policies to AI Overviews and AI Mode. The way to get cited is still extractable revenue pages, entity corroboration, and freshness, not a page for every fan-out query. That playbook is unchanged.