Koka Sexton

TL;DR

• Google’s AI Overviews now appear on 47% of search queries, and more than 60% of all Google searches generate zero clicks to any website.
• ChatGPT, Perplexity, and Claude have become the default research layer for B2B buyers. They cite sources based on conceptual authority, not domain authority.
• Publishing more SEO-optimized content is the wrong move. The volume play is dead, and the velocity play is dying.
• The winning strategy is building an AI-native content moat: fewer pieces, deeper concept ownership, and distribution optimized for LLM citation rather than keyword ranking.
• Brands that restructure their content strategy for the AI search era now will own the next five years of B2B discovery. Everyone else will be invisible at the moment of highest buyer intent.

---

The Traffic You’re Chasing Is Already Gone

Let’s get specific about what’s happening, because most of the coverage on AI search is either vague or panicked. Neither is useful.

SparkToro’s zero-click search research has tracked a steady rise in queries that end without a click to any website. By 2025, that number crossed 60% for the first time. More than six out of ten Google searches now resolve entirely within Google’s ecosystem: AI Overviews, knowledge panels, featured snippets, and People Also Ask boxes.

That number was already climbing before AI Overviews went mainstream. Now it’s accelerating.

Google’s AI Overviews appear on 47% of search queries as of mid-2026, according to multiple third-party tracking studies including BrightEdge and Authoritas. For commercial queries – the kind B2B buyers run when evaluating solutions – the appearance rate is even higher. Those searches involve comparison and synthesis that AI models handle well.

Here’s what that means in practical terms.

A CTO searching “best CRM for mid-market SaaS” used to land on a comparison page, read it, maybe clicking through to two or three vendor sites. Today, Google’s AI Overview summarizes the top options, lists key features, and often recommends a shortlist – all without the searcher leaving the results page. Perplexity does the same thing with citations. ChatGPT does it conversationally.

The content that used to capture those high-intent visitors is still being published. It’s just not being visited.

This is the core dynamic most content teams haven’t internalized: the content is still there. The buyer intent is still there. But the connection between them – the click – is being systematically removed.

SEO Was Always a Distribution Channel. Now It’s Shrinking.

The mistake most content strategists make is treating SEO as the purpose of content rather than one of its distribution channels.

SEO isn’t dead. But as a primary distribution mechanism for B2B content, it’s in structural decline. Not because Google is getting worse at search; it’s getting better. It’s getting so good at answering questions that it doesn’t need to send people anywhere to get the answer.

This distinction matters because it reframes the problem. You’re not fighting a Google algorithm update. You’re adapting to a platform-level shift in how information reaches buyers.

Consider what AI search actually does with content.

It synthesizes, it doesn’t rank. Google’s traditional search ranked pages by relevance signals and sent you there. AI search reads multiple sources, synthesizes an answer from them, and presents it as a finished product. The page that wins isn’t the one with the best SEO. It’s the one the AI trusts as the most authoritative source on that specific concept.

It cites concepts, not keywords. When Perplexity or ChatGPT cites a source, it’s not matching keywords. It’s evaluating whether the source credibly owns the concept being discussed. A blog post that ranks #1 for “B2B lead scoring” gets zero AI citations if it’s generic and surface-level. A less-optimized post that defines a specific framework with clarity and evidence gets cited repeatedly.

It rewards differentiation, not comprehensiveness. Traditional SEO rewarded the post that covered everything. AI search rewards the post that covers one thing definitively. The 4,000-word “Ultimate Guide to B2B Marketing” gets ignored. The 1,800-word piece on why your lead scoring model breaks at scale gets cited.

A Meltwater study published in May 2026 analyzed 9.5 million citations across ChatGPT, Perplexity, and Google’s AI Overviews. The finding was stark: AI engines overwhelmingly favor content that demonstrates clear conceptual ownership – frameworks, methodologies, and data-backed analysis – over content optimized for keyword breadth.

This completely inverts the content strategy most B2B teams are running.

The Content Factory Is Producing Invisible Product

Walk through the logic of a typical B2B content operation today.

You have a content calendar. You publish weekly, sometimes twice weekly. Each piece targets a keyword cluster, follows a templated structure, includes internal links, ends with a CTA. The goal is volume: more pages, more keywords, more surface area for Google to index.

This model worked when Google sent traffic proportionally to ranking authority. It doesn’t work when Google answers the question itself.

The average B2B blog post now competes with three answer engines that don’t need to send traffic anywhere. It competes with Google’s own answer, Perplexity’s synthesized response, and whatever ChatGPT tells the buyer when they ask. All three are getting better at answering without citing anyone at all.

Here’s the uncomfortable math.

A content team producing two posts per week – call it 100 posts per year – spends $150,000 to $300,000 annually on content creation, depending on quality and team structure. If organic traffic declines 30-40% over the next two years, as multiple forecasts project, that team is spending the same money for dramatically less reach.

The obvious response – publish more to compensate – is exactly wrong. More content in a shrinking channel means more competition for fewer clicks. The volume play is dead. The velocity play is dying.

The AI-Native Content Moat

Here’s a framework for content strategy in the AI search era. I call it the AI-Native Content Moat.

The premise is simple: build content that AI search engines can’t afford to ignore. Not because you gamed their algorithm, but because your content owns concepts so clearly that any AI synthesis on the topic feels incomplete without citing you.

The moat has three layers.

Layer One: Concept Ownership

Most B2B content borrows concepts from other content. It rephrases the same frameworks, cites the same studies, reaches the same conclusions. AI search engines see this and treat it as interchangeable. Any one of ten similar articles can supply the answer.

Concept ownership means defining something original. Not necessarily a new idea – that’s rare – but a specific, named articulation of an idea that becomes the reference point for discussions of that topic.

Think about how HubSpot owns “inbound marketing” or how Drift owned “conversational marketing.” Those aren’t just content strategies. They’re concept strategies. When an AI model encounters the phrase “inbound marketing,” it reaches for HubSpot. When a buyer asks Perplexity about conversational marketing, Drift appears in the citations. That’s not SEO. That’s entity authority.

Building concept ownership requires a deliberate shift from topic coverage to concept definition. Instead of writing a post about “B2B demand generation,” you write a post that defines a specific methodology for B2B demand generation. Name it, structure it, illustrate it with data, link to it from every subsequent piece.

One strong concept owns more AI search real estate than fifty generic blog posts.

Layer Two: Citation Architecture

AI search engines don’t just read your content. They read your content’s structure: its schema markup, its heading hierarchy, its entity relationships, its internal linking patterns.

Citation architecture is the deliberate structuring of content to make it maximally citable by AI models.

This isn’t traditional on-page SEO – there’s meaningful overlap – but the goal is different. You’re building content that an AI can parse, extract from, and attribute confidently in a single pass.

The key elements:

Entity-rich content. Every piece you publish should clearly identify the entities it discusses: companies, people, frameworks, products, categories. AI models build knowledge graphs from entity relationships. Content that makes those relationships explicit gets surfaced more often.

Self-contained sections. AI models extract insights by section, not by page. Each H2 section of your content should be understandable in isolation, containing its own context, its own evidence, and its own conclusion. When an AI pulls a paragraph from your article as a citation, that paragraph alone should make sense and point back to your framework.

Schema that speaks AI, not just Google. FAQ and HowTo schema remain critical because they feed directly into AI-generated answers. But Article schema with properly populated about and mentions properties is increasingly important. It tells AI models which concepts your content addresses and which entities it references, enabling more accurate attribution.

Consistent conceptual language. If you call something “revenue architecture” in one post and “revenue operations” in another, AI models treat them as separate concepts. Pick your terms and use them consistently across every piece of content. Every piece should reinforce the same entity map.

Layer Three: Distribution Density

The final layer of the moat is distribution density: getting your concepts cited in enough places that AI models encounter them from multiple independent sources.

AI models trust concepts they see referenced repeatedly across different domains. One great article on your blog is decent. That same article cited by three industry publications, referenced in a LinkedIn thought-leader post, and discussed on a respected podcast – now the AI treats it as consensus.

This doesn’t require a PR agency. It requires a distribution strategy built around getting your concepts placed in the publication ecosystem:

• Write for industry publications using your named frameworks
• Appear on podcasts and reference your concepts naturally
• Build LinkedIn content that reinforces your core ideas (more on this in my piece on LinkedIn content strategy in a saturated market)
• Create original research that other publications will cite

The goal isn’t backlinks for SEO. It’s citation density for AI authority. When six different domains reference your framework by name, AI models treat it as established knowledge.

The Counterintuitive Move: Publishing Less, Owning More

If you take one thing from this piece, take this: the right response to AI search is to publish fewer pieces of content, not more.

This sounds heretical to anyone running a content calendar. But the logic is straightforward.

AI search engines don’t reward volume. They reward authority. And you don’t build authority by publishing 100 surface-level posts. You build it by publishing 20 deep, conceptually owned pieces that AI models cite repeatedly.

The content teams winning the AI search era are already doing this. They’re cutting their publishing cadence and reinvesting that time into:

• Deepening existing content: adding original data, sharper frameworks, clearer concept definitions
• Building distribution density: getting their concepts cited across the publication ecosystem
• Monitoring and improving AI citations: tracking which pieces get surfaced in AI answers and strengthening the ones that don’t

The tool consolidation math I cover in my martech sprawl audit applies here too. An AI orchestration layer replaces the point solutions your content team is maintaining. Fewer tools, deeper output, better AI citability. The same consolidation logic that cuts SaaS waste also produces stronger content.

The irony is that this produces better human outcomes too. Buyers who encounter a brand through an AI citation arrive with more trust and higher intent than buyers who clicked a Google result. The AI citation functions as a third-party endorsement.

A study by Conductor found that brands cited in AI Overviews saw 40% higher brand recall than brands appearing in traditional search results. The AI citation carries a credibility signal that a search ranking doesn’t.

What to Do This Quarter

This isn’t a think piece about the distant future. It’s a playbook for the current quarter. Here’s what to do.

The brands that execute this shift now will own B2B search for the next five years. Everyone else will keep publishing weekly blog posts into a distribution channel that’s actively shrinking, wondering why the traffic keeps falling but the calendar never changes.

The Obstacle: Your Metrics Are Stuck in the Old Model

It’s worth naming the obstacle, because it’s predictable.

Most content teams know something is wrong. They can feel the traffic decline, sense the platform shift, see their competitors experimenting with new formats. But they can’t change because their content operation is measured by the wrong metrics.

If your CEO measures content success by monthly publish count and organic traffic growth, you’ll keep publishing more and watching traffic decline. The metrics themselves are stuck in the old model.

The fix starts with reframing the conversation. Stop reporting publish volume and start reporting concept ownership. Track AI citation frequency instead of, or alongside, keyword rankings. Measure how often your frameworks appear in AI-generated answers, not how many pages you published this month.

The content teams making this shift successfully are the ones whose leaders understand that the game has changed. If your leadership doesn’t get it yet, send them this article. The data is clear. The playbook is here. The only question is timing.

---

Build Content AI Search Engines Can’t Ignore

The AI search shift isn’t coming. It’s here. The content playbook that built B2B marketing over the last decade was designed for a version of the internet where Google sent traffic to websites. That version is ending, and the teams that adapt their strategy now will own the discovery channels their buyers actually use for the next five years.

Let’s build a content strategy that earns citations, owns concepts, and drives pipeline through AI-native discovery.

Let’s Talk Strategy