Author: Bill Ross | Published: February 4, 2026 | Updated: May 24, 2026 Modern entity-based SEO treats your brand, your people, your products, and your topics as connected nodes in a machine-readable graph, not as a list of keywords waiting to be matched. Google’s Knowledge Graph holds 54 billion entities. AI Overviews now appear on roughly 48% of tracked queries. Click-through rates at position one have fallen 58% when an AI summary is present. The work that wins organic visibility in 2026 looks nothing like the work that won in 2018, and the brands seeing the cleanest results are the ones who stopped optimizing for “strings” and started optimizing for “things.” Key takeaways: An entity is a thing that can be uniquely identified. A person, a company, a product, a place, a concept. Google’s Knowledge Graph treats each of these as a node with attributes and relationships, not as a string of letters that might appear on a page. When you search for “Apple,” Google decides whether you mean the company, the fruit, or the band based on context, then surfaces information about that specific entity. The graph started small. In December 2012, Google reported 570 million entities. By May 2024, that figure had climbed to roughly 54 billion. In June 2025, Google deleted more than three billion entities in what the search community now calls the “clarity cleanup,” signaling a shift from volume to confidence. The graph is now growing more carefully, with each entity assigned a clearer, more specific type. This matters because Google’s AI features, including AI Overviews and AI Mode, are built on top of this graph. If your brand exists as a clear entity, with attributes (founders, services, locations, products) and relationships (industries served, partners, certifications), you become eligible for citation. If you exist only as a website with good copy, you may not.
“The brands we see winning right now aren’t producing more content. They’re producing content that reinforces a clearer identity. Every page answers the same question: who is this entity and what does it know? The graph picks that up.” – Emulent Strategy Team
The shift from string-matching to entity-matching began with Hummingbird in 2013, deepened with RankBrain in 2015, and accelerated with BERT in 2019. AI Overviews are the consumer-facing endpoint of more than a decade of semantic work. Which raises the obvious question: what does that decade of work do to traditional keyword-first SEO? The clearest answer is in the data. Google’s AI Overviews moved from edge case to default in roughly 24 months. In January 2025, AI Overviews appeared on 6.5% of tracked Google queries. By February 2026, that share reached 48% according to BrightEdge. The peak in July 2025 was higher still, around 24.6% on Semrush’s tracked set before Google pulled back to clean up quality issues, then pushed coverage forward again. The consequence for ranked links is measurable. Ahrefs analyzed 300,000 Search Console keywords in December 2025 and found that the top-ranked organic result loses 58% of its clicks when an AI Overview sits above it. Position two loses about 51%. Position three loses roughly 34%. The clicks that survive go disproportionately to brands cited inside the Overview itself, which earn 35% more organic clicks than uncited brands at the same position. What this changes for an SEO program: The CTR collapse is not the whole story. It only counts users who still see a SERP. A growing share of searches don’t even reach that step. Two trend lines define the next few years. The first is zero-click search. SparkToro and Datos first quantified the phenomenon in 2019, when roughly half of Google queries ended without a click to an external site. By 2026, that share has climbed to roughly 65%. Some queries (a definition, a currency conversion, a weather forecast) never needed a click. Others used to drive traffic and now stop on the SERP.
“AI traffic looks small until you measure quality. The conversion rate on a ChatGPT referral runs four to twenty-three times the rate of a Google organic visit, depending on the vertical. A 1% traffic channel can drive 10% of revenue. That changes the math on where you invest.” – Emulent Strategy Team
The combined picture is straightforward. Total search activity is rising, but the share that ends in a website visit is falling, while a new class of visit (from AI answer engines) is rising fast. Winning under those conditions requires being legible to both Google’s classical ranking systems and the answer engines that now sit on top of them. Which raises the next question: what makes a brand legible? Three signals do most of the work. Structured data on your own site. Consistent identity across the open web. And mentions on sources Google already trusts. Each of these gives the Knowledge Graph a piece of confirmation. None of them alone is enough. Structured data is the most controllable. Schema.org markup tells Google exactly what your page is about, who wrote it, what organization stands behind it, and how it relates to other entities. Adoption has climbed steadily for a decade. In 2010, only 5.7% of web pages used any form of structured data. By 2024, that figure had reached 51.25%. Of pages ranking in Google’s top ten, 72.6% already use schema. The structured data layer that supports entity recognition: Beyond your site, identity consistency carries weight. Your name, address, phone, founders, and category should match across Google Business Profile, LinkedIn, Crunchbase, Wikidata, and industry directories. Ambiguity kills citations, as we’ve seen in our AI SEO engagements. If three sources say your headquarters is in Wake Forest and one says Raleigh, the graph hesitates. If the same employee shows three different job titles across LinkedIn, your bio page, and a podcast description, the graph picks the highest-confidence option and ignores the others. That brings us to the third signal: who mentions you, and what they say. This is where entity SEO begins to look less like SEO and more like classical brand work. Several research groups have measured citation drivers in AI Overviews, ChatGPT, Perplexity, and Gemini. The ranked picture is consistent across studies. Branded web mentions and earned media are the largest levers. Schema and structure are necessary but smaller. Page-level tactics like comparison sections and recency help but don’t carry the whole load. The pattern matters because it inverts where most marketing teams spend their SEO budget. Most invest heavily in on-page content production and lightly in distribution. The data says distribution and earned mentions return more visibility per dollar than another keyword-optimized page on a brand’s own blog. A brand strategy that produces quotable points of view, original research, and contributed work on trusted publications now feeds the SEO machine more directly than it did five years ago.
“We tell clients to stop counting their published pages and start counting the publications that quote them. One credible third-party citation in an industry outlet does more for AI visibility than ten more posts on the company blog.” – Emulent Strategy Team
The earned-media playbook for entity authority: If you’re tracking the right things, you can see entity SEO working before it changes your rankings dashboard. So how should you measure it? Traditional SEO reporting tracks keyword positions, clicks, impressions, and conversions. Those metrics still matter, but they describe a shrinking part of the picture. An entity-based program needs three additional layers of measurement. The metrics that matter under entity-based SEO: The reason these metrics matter is that they catch movement that ranking reports will miss. A brand can gain real visibility for months before that visibility shows up in traditional click data, because AI Overviews absorb the early gains. Teams that only watch rankings will conclude their program isn’t working when it actually is. Teams that watch citations and share of voice see the lift early and can double down. A practical companion piece is our AI SEO checklist, which walks through the technical and editorial work in order. For the broader playbook on optimizing across Google, AI search, social, and answer engines together, see our work on search everywhere optimization. The shift to entity-based SEO is not a tactic. It is a different way of thinking about what your brand is to a machine, and the work cuts across content, technical SEO, public relations, and brand strategy. We help organizations rebuild around that model, from auditing your Knowledge Graph presence and structured data, to producing the earned media that drives citation lift, to setting up the reporting infrastructure that catches AI visibility before it shows up in revenue. If you’re seeing impressions rise while clicks fall, watching competitors get cited in AI Overviews while your better content sits below them, or trying to figure out how to measure success when ranking position no longer maps to outcomes, we can help. Contact our team to talk about how entity-based SEO and AI search optimization fit into your next quarter.
A Guide to Entity-Based SEO That Goes Beyond SEO Keywords
What does “entity-based SEO” actually mean today?
Why is keyword-first SEO losing leverage?
Where is search behavior actually heading?
The second trend line is AI referral traffic. ChatGPT, Perplexity, Gemini, and Copilot now send measurable referral visits to websites. Volume grew 357% year over year in 2025, with AI platforms sending roughly 1.13 billion referral visits in June 2025 alone. The channel is small in absolute terms (somewhere between 1% and 3% of total referral traffic at present), but the trajectory is steep and the visits convert.
How does Google identify your brand as a real entity?
What actually earns a citation in AI answer engines?
How should you measure entity SEO success?
How Emulent can help
