Stop Making More Content: The New Rules of AI Search Visibility

Here’s a sentence I never thought Google would write: stop making so much content. But that’s essentially what Google just published in its first official guide on optimizing for AI Search. And it lands in the same week that new benchmark data shows publisher ad volume dropping by up to 40% because AI answers the question before anyone clicks. Something big is shifting, and if your 2026 marketing plan still says „publish more, add more schema, chase more mentions,“ you’re planning for a game that no longer exists.

For twenty years the playbook was simple: more content, more keywords, more backlinks, and eventually you win. AI search broke that math. The channels are consolidating, the clicks are disappearing, and the machines doing the recommending have very different taste than the algorithm you spent a decade optimizing for. This is the moment to rethink how your brand stays visible.

Below is what changed, what the data actually says, and the specific moves that still work — pulled from Google’s own guidance, a 20-billion-impression publisher study, and a four-month experiment that ran nearly 10,000 AI answers through the wringer.

Google Just Told Marketers to Do Less

In its first guide on optimizing for AI Search, Google warns against three habits that most content teams treat as gospel: creating dozens of near-identical articles, over-investing in structured data, and chasing inauthentic brand mentions. Instead it keeps circling back to one idea — create „non-commodity content.“

The logic is brutal and simple. If your content says the same thing as everyone else’s, an AI engine has no reason to cite you. Neither does Google. And most companies badly underestimate how commoditized their own library has become. It’s optimized for keywords but adds nothing new: no original data, no unique insight, no firsthand experience.

If your content says the same thing as everyone else’s, AI has no reason to cite you. Neither does Google.

That’s a profound reframe for anyone still measuring content marketing by output volume. The metric that matters now isn’t how many posts you shipped — it’s whether any single one of them says something the internet couldn’t already answer without you.

The Open Web Is Drying Up — Fast

If you needed a reason to act now, here it is. New benchmark data from Ozone (run across roughly 20 billion impressions for premium publishers like The Guardian, News UK and The Wall Street Journal) shows the open web contracting hard as AI answers absorb the clicks.

When the AI delivers the answer inside search, nobody clicks through. No click, no page view. No page view, no ad slot. Revenue is falling slower than volume only because prices are climbing — UK eCPMs sat about 30% above last year — but that’s a temporary cushion, not a rescue. The strategic takeaway for marketers is blunt: buying broad visibility through big channels is getting more expensive and less reliable every quarter.

Self-Promotion Works — Until It Backfires

So if paid reach is shrinking, can you just publish your way into the AI’s answers? Sometimes. Ahrefs ran a sharp four-month test: they published 34 promotional pages across five domains — one for a brand-new conference, one for an established tool — then analyzed nearly 10,000 AI answers from ChatGPT, Gemini, Perplexity and Copilot.

The results split cleanly by brand maturity. For the new brand, self-promotion worked: the conference showed up in 72 previously-empty answer slots, and 82% of those new mentions came from answers that cited its own pages. For the established brand, it barely moved: only 6% of new mentions came from its own pages, while 94% came from third-party sources. The AI already had plenty of material and didn’t need the brand’s own marketing.

Two more findings every CMO should tattoo on the wall. First, context decides everything: the conference appeared in 66.4% of answers to „best SEO conferences 2026“ but just 15.8% of answers to „best marketing conferences 2026.“ Relevance to the exact question mattered four times more than raw presence. Second, citations are fragile: between a page’s first and last mention it was cited on only about one in three possible days, and roughly a quarter of pages appeared once and then vanished.

The New Playbook for 2026

Put the three signals together and a clear strategy falls out. A Semrush survey of 480+ marketers found most have already folded AI search into their strategy — the gap now isn’t awareness, it’s execution. Here’s where to point that execution.

Practically, that means fewer briefs and bigger ones. Pull the numbers only you have — your customer data, your benchmarks, your results — and build content around them. Take a real position instead of hedging; AI engines prefer clear, definitive statements they can extract. And treat your own team as a distribution channel: firsthand expertise from named humans reads as exactly the „non-commodity“ signal Google is asking for.

Frequently Asked Questions

What is „non-commodity content“?

Non-commodity content is material that adds something the internet doesn’t already have — original data, unique insights, or firsthand expertise. Google’s AI Search guidance singles it out as the main thing worth creating, because AI engines have no reason to cite content that simply restates what’s already everywhere.

Is SEO dead in the AI search era?

No, but the „more content“ version of it is fading. Fundamentals like clear structure, relevance and authority still matter. What’s changing is that volume and keyword-stuffing no longer win — originality, specific intent and citability do.

Why are publisher ad revenues dropping?

Because zero-click AI search answers questions inside the results page, so users don’t click through. Fewer clicks mean fewer page views and fewer ad slots. Q2 2026 benchmark data showed ad volume down up to 40% year over year in the US and UK.

Can I just publish promotional pages to get cited by AI?

Only if you’re filling a genuine gap. In a four-month test, a new brand grabbed 72 empty answer slots this way, but an established brand saw just 6% of new mentions come from its own pages. Self-promotion amplifies credibility you already have — it doesn’t manufacture it.

What’s the single most important shift for CMOs?

Move from measuring output to measuring influence. Publish fewer, sharper pieces built on data and expertise only you can offer, own the specific questions in your category, and track AI citations continuously because they’re volatile.

Foto von Igor Omilaev
auf Unsplash