If that sentence feels familiar, it should.
The September 2026 spam update began on September 24 and is the fourth confirmed spam update Google has released this year, following updates in March, June, and August.
This one comes with an important difference.
Google says the rollout may take up to two weeks.
March finished in roughly 19 hours. June took a little over two days. August finished in 2 days and 16 hours. Google’s John Mueller has confirmed that the longer September window is intentional and that this update is likely to take longer than some of the previous ones.
So, for once, automotive marketers have been given an excellent reason not to draw conclusions from the dashboard before Monday morning.
As of this writing, the update is still underway.
That qualification matters. There is already considerable volatility showing across third-party tracking tools and industry chatter, including movement immediately before the announcement and another increase after rollout began. But the next several days are likely to contain enough ranking movement, reversals, partial recoveries, and unexplained weirdness that responsible analysis needs to remain just that: analysis.
Google has not announced a new spam policy. It has not said this is an AI-content update. It has not disclosed what specific systems or tactics received additional attention.
What it has done is launch its fourth spam update in six months while giving this one a substantially longer runway.
For retail automotive, that is worth more than another afternoon staring at a volatility chart.
Google released the September 2026 spam update on September 24 at 9:15 a.m. Pacific.
According to the Google Search Status Dashboard, it applies globally and across all languages. Google says the rollout may take up to two weeks to complete.
That makes it the fourth confirmed spam update of 2026.
Google has not announced a new category of spam or published a special September rulebook. The existing Google Search spam policies remain the relevant standard.
Google describes spam updates as notable improvements to automated systems that operate continuously to detect Search spam. SpamBrain, Google’s AI-based spam-prevention system, is one example of those systems.
If a site changes materially during a spam update, Google recommends reviewing its spam policies. Sites found violating those policies can rank lower or disappear from results, and recovery can take months while Google’s systems relearn that the site is compliant.
There is no five-minute recovery button hiding in Search Console.
The most interesting factual difference between September and the three earlier 2026 spam updates is the expected duration.
March took less than a day.
June took just over two.
August took under three.
September may take two weeks.
That does not automatically tell us the update is “bigger.” Google has not said that. Rollout duration is not a severity score.
But operationally, it does change how dealerships should handle the next several days.
A longer rollout creates more room for rankings to move at different times across queries, sites, languages, markets, and page classes. That makes premature diagnosis even more dangerous than usual.
If a service page drops Wednesday and returns Saturday, rewriting it Thursday afternoon would have taught you very little.
If one rooftop in a group moves before another, that does not necessarily mean the stores were evaluated differently in some final sense. They may simply be moving through the rollout at different times.
And if a dashboard looks heroic at breakfast and terrible by dinner, congratulations: you have encountered Google Search in 2026.
During this rollout, patience is not passivity. It is measurement discipline.
This part is important because the industry has already started attaching every September chart to the newly announced spam update.
Search Engine Roundtable documented substantial ranking volatility around September 23 and 24, with some signals beginning on September 22. There had also been separate volatility around September 15 and unusual movement earlier in the month.
The timing gets messy because the confirmed spam update did not begin until September 24 at 9:15 a.m. Pacific.
Some of the movement immediately around launch may ultimately overlap with the rollout. Some of the earlier September movement plainly predates it.
So dealerships should resist turning “September was weird” into “the spam update did it.”
Those are not the same statement.
When the rollout concludes, good analysis should distinguish:
A red line in a reporting platform tells you something changed.
It does not tell you why.
March.
June.
August.
September.
Google has now launched four confirmed spam updates in roughly six months.
We should not invent intent Google has not stated. There is no public announcement saying this cadence represents a new enforcement campaign, a war on AI content, or a specific crackdown on automotive publishing.
But automotive executives should pay attention to the environment those updates are occurring inside.
The cost of content production has collapsed.
Every agency has AI.
Every website provider can generate pages.
Every SEO tool can find another keyword.
Every vehicle feed contains thousands of structured attributes.
Every dealership has dozens of geographic, service, vehicle, trim, ownership, financing, comparison, and FAQ combinations available to turn into URLs.
The publishing bottleneck is disappearing.
Google’s abuse-detection systems are clearly not standing still while that happens.
This creates an important change in the economics of automotive content.
Being able to produce content at scale is no longer an advantage by itself.
Almost everyone can do that now.
The advantage increasingly belongs to the organization that can decide what deserves to be produced, ground it correctly, govern it responsibly, connect it intelligently, and make it meaningfully better than the commodity alternatives.
Retail automotive may be one of the easiest industries on earth to programmatically create content for.
We have structured inventory.
OEM specifications.
Trim data.
Incentives.
Finance programs.
Service intervals.
Parts catalogs.
Recall information.
Geography.
Dealer groups.
Thousands of combinations of all of the above.
A capable system can take that data and generate an astonishing amount of content.
The question Google’s policies force us to ask is whether we are adding value or simply adding URLs.
Google defines scaled content abuse as generating many pages primarily to manipulate search rankings rather than help users. Its examples explicitly include using generative AI to create large amounts of low-value content, transforming feeds without adding meaningful contribution, combining material from other sources without improving it, and creating pages largely around search keywords rather than human usefulness.
Again, this does not mean automation is bad.
It means automation is accountable for its output.
A page built from a clean OEM feed can still be useless.
A page created with an LLM can be outstanding.
A page written entirely by a human can be spammy.
The production method is not the interesting part anymore.
The contribution is.
This distinction remains badly needed in automotive.
Google does not prohibit AI-assisted content.
Its own guidance acknowledges legitimate uses of generative AI for research, organization, and content creation. The problem arises when automation is used to create large amounts of content without adding enough value for the user.
That makes AI less of a content question and more of an operating-model question.
Imagine a technician who has worked on hundreds of F-150s explaining the symptoms that usually precede a battery failure.
AI helps structure that knowledge, match it to relevant vehicle data, organize the article, generate supporting FAQs, connect it to related service resources, and keep the finished page within brand and compliance standards.
That is intelligent leverage.
Now imagine taking the phrase “Ford F-150 battery service” and automatically creating hundreds of city pages with slightly different introductions.
That is also AI-assisted content.
The resemblance mostly ends there.
One system is scaling expertise.
The other is scaling inventory.
Our industry is going to spend the next several years learning that those are not remotely the same business.
The rise of AI search has created a new temptation.
Google’s AI systems can use query fan-out to explore multiple related questions before constructing an answer. Marketers understandably see those possible subqueries and start thinking about coverage.
Then somebody opens a spreadsheet.
This is usually where things get interesting.
Google’s own guidance for generative AI features now explicitly warns against creating separate pages for every possible query variation when the primary purpose is manipulating rankings or generative AI responses. Google points directly back to its scaled content abuse policy.
This should matter enormously to automotive.
The answer to conversational search is not to create one URL for every sentence a customer might type.
It is to create stronger bodies of knowledge capable of answering related questions coherently.
We explored this recently in The Next Question Is the New Search.
A buyer rarely stops at:
“What is the best three-row SUV?”
They continue.
Which one has enough room behind the third row?
Which is best for car seats?
What does the hybrid cost to maintain?
Can it tow my camper?
Do you have one?
Can I drive it Saturday?
The dealership that understands that conversation does not need 900 disconnected pages.
It needs a connected knowledge system capable of staying useful as the customer keeps asking better questions.
The strongest answer to scaled-content risk is not “publish less.”
It is “know more about what you publish.”
Dealerships are sitting on a remarkable amount of proprietary context:
The problem is that most of this information lives in separate systems, separate departments, or people’s heads.
We have been writing extensively about this through Hrizn’s recent work on the dealership intelligence layer and knowledge graph.
The shift matters for search because generic content becomes much harder to justify when a dealership can instead publish something grounded in what it actually knows.
That is the content competitors cannot reproduce by changing a prompt.
The dealership’s own knowledge becomes the moat.
If dealerships want a practical example, walk into the service department.
The technicians, advisors, and fixed ops leaders inside most stores collectively answer hundreds of real customer questions every week.
They know which issues are common.
They know which sounds customers describe incorrectly.
They know what extreme heat does to batteries.
They know what road salt does to vehicles.
They know when a warning light is urgent and when it is merely irritating.
They know which maintenance customers routinely misunderstand.
And yet many dealership websites reduce all of that accumulated knowledge to:
“Need brake service? Our factory-trained technicians are here to help.”
Technically true.
Also not exactly Pulitzer material.
A stronger approach turns firsthand service expertise into structured ownership content. Hrizn’s Fixed Ops Content Marketing framework is built around exactly this opportunity.
That content can support search rankings, AI citations, local visibility, service retention, trust, and actual appointments.
Most importantly, it has a reason to exist.
For dealer groups, the September rollout should be analyzed at two levels.
The first is the rooftop.
The second is the system.
If one dealership loses visibility on a page, investigate the page.
If 14 rooftops lose visibility across the same content template, investigate the architecture.
Group operators should look for patterns across:
This is where enterprise-level measurement becomes much more useful than one-store reporting.
A group can see whether the issue follows a market, a brand, a provider, a template, or a publishing methodology.
That is actionable intelligence.
OEMs have a similar challenge at much larger scale.
Manufacturers shape dealership content through compliance rules, syndicated data, approved technologies, website standards, incentives, national campaigns, and shared content programs.
At a few thousand rooftops, a mediocre content decision becomes infrastructure quickly.
OEM teams should be asking whether their systems preserve useful local differentiation or slowly sand it away.
Can a dealer combine authoritative OEM data with firsthand local expertise?
Can a service director contribute useful knowledge without creating compliance chaos?
Can national product information remain authoritative while regional and dealership context improves it?
Can machine-generated content carry clear provenance?
Can outdated information be corrected once and propagate reliably?
Can the OEM see which content systems are creating useful visibility across the network and which are simply generating pages?
These are increasingly search questions, AI questions, governance questions, and infrastructure questions at the same time.
Good governance is difficult when the underlying infrastructure fights you.
Dealerships operate across website providers, CRMs, inventory systems, analytics platforms, content tools, social networks, OEM systems, and increasingly custom AI agents.
If those systems cannot exchange data cleanly, marketing teams compensate manually.
They copy.
They paste.
They duplicate.
They create workaround landing pages.
They automate browsers because there is no usable API.
Then everybody wonders why nobody can explain where half the content came from.
Modern content governance needs better plumbing.
Open APIs, controlled CRUD access, webhooks, clear permissions, structured data, and reliable interoperability make it easier to scale useful content without losing provenance.
This has been a recurring Hrizn argument for a reason.
Responsible scale is not achieved by avoiding automation.
It is achieved by building infrastructure that makes automation governable.
Mostly?
Keep your hands off the big red button.
The September spam update may continue rolling for up to two weeks. Major reactive changes during the middle of that process can make later analysis considerably harder.
That does not mean doing nothing.
Use the rollout window to establish your baseline.
Preserve page and query performance for the weeks before September 24. You will want clean comparison periods later.
Build page groups for service, model research, local pages, blog/editorial, inventory-supporting content, and major programmatic templates.
Know which vendors, integrations, plugins, and automation systems are currently capable of adding or changing content on the domain.
If you discover thousands of low-value doorway pages, that deserves attention regardless of the update. But distinguish an existing governance problem from a proven September-update impact.
One ranking moving from four to seven is not strategic intelligence.
Fifty related service queries moving across ten similar pages might be.
Once Google marks the update complete and the immediate movement settles, compare clean pre- and post-update periods.
Did service, research, editorial, local, or programmatic content move disproportionately?
Separate branded searches from discovery searches. Branded demand can hide a lot of organic weakness.
If pages using one common structure move together, investigate the structure rather than rewriting every page independently.
Where did losing content come from?
OEM feeds?
Legacy agency programs?
Generic AI generation?
Dealership experts?
Third-party syndication?
The answer may expose an operating-model issue.
Do not stop at impressions and clicks.
Which pages actually contributed to appointments, inventory engagement, lead activity, and useful customer journeys?
A page with modest traffic and high service conversion may deserve far more protection than a generic blog article producing ten times the sessions and zero business.
And please do not respond to a spam update by publishing a press release explaining that your dealership is committed to authentic content.
Just make the content authentic.
There will be plenty of claims over the next several days about what the September update “targeted.”
Some will eventually prove useful.
Some will be based on legitimate data.
Some will be a screenshot of three websites and a confident LinkedIn caption.
Hrizn will continue watching the rollout and will publish a full post-update analysis once Google marks it complete and there is enough stable data to say something useful.
Until then, the productive takeaway for automotive is less dramatic.
Publishing capacity has become cheap.
Knowledge has not.
Judgment has not.
Trust has not.
And a dealership’s ability to turn firsthand expertise into connected, governed, machine-readable content is becoming more valuable as every competitor gains access to the same generation tools.
That is a much more interesting advantage than figuring out how to publish another thousand pages before Google finishes rolling this thing out.
Continue exploring Hrizn:
Primary references:
We Rise Together.
