Build memorable influence across every channel by diversifying your brand presence for organic and AI success.
The rules of digital visibility have changed — and they’re not changing back. If your entire brand strategy still revolves around ranking blue links on a Google results page, you’re building on a shrinking foundation. The smartest brands right now are rethinking how they show up everywhere: in search engines, in AI-generated answers, in social feeds, in newsletters, in podcasts, and in the minds of people who never clicked a link to find them. That’s what it means to build a memorable influence in 2026 and beyond.
Here’s what’s actually happening to organic search, why AI is reshaping how people find information, and what you can do right now to keep your brand visible, trusted, and cited across every channel that matters.
Stop optimizing for Google, start building a brand AI can’t ignore. Google stopped sending traffic, they just didn’t tell you - 64% of google searches never click a single link.
Start with the data, because it’s genuinely sobering. Research from SparkToro and Datos found that for every 1,000 U.S. Google searches, only about 360 clicks go to the open web. The remaining 640 either produce zero clicks or land users on another Google-owned property — YouTube, Maps, Images, or an AI Overview. Extrapolated across Google’s estimated 5 trillion annual searches, roughly 3.2 trillion searches never reach a third-party website at all.
That’s not a blip. That’s structural.
Google’s own internal strategy documents — surfaced as evidence in ongoing antitrust litigation — showed executives weighing three futures for their search product: traffic holds steady, users migrate to Gemini, or users migrate to ChatGPT. Their preference was obvious: keep traffic inside Google’s walls. The technology to actually do that now exists in a way it simply didn’t five years ago. That changes the calculus for every publisher, brand, and content creator who has depended on Google to send people their way.
Gartner predicted in 2024 that traditional search engine traffic will drop 25% by 2026. A lot of people in the SEO world shrugged it off. Fewer are doing that now. Cloudflare CEO Matthew Prince has been publicly critical of what he describes as AI platforms consuming web content to generate answers and then keeping users on-platform — a behavior he argues is quietly dismantling the economic model that made the open web worth building.
Understanding this context isn’t an exercise in pessimism. It’s the prerequisite for making clear-eyed decisions about where to put your energy.
Here’s where that definition gets interesting.
Brand mentions used to be a brand awareness term — basically, can people recall your name after an ad impression? Now it means something more specific and more technically important: does your brand exist as a recognized, citable entity in the information ecosystem that AI models, search engines, and human audiences are all drawing from simultaneously?
Google laid out the company’s long-term vision as far back as 2008 — build the Star Trek computer, the one that gives you a direct answer rather than a list of links to go sort through yourself. DeepMind co-founder Mustafa Suleyman put a date on it in more recent commentary: by 2033, conversation will be the default interface for information retrieval. Not a search box.
That transition is happening faster than anyone predicted. Google’s AI Overviews are already appearing in the first organic position on millions of queries. AI Mode is in active testing. ChatGPT, Perplexity, Claude, Gemini, and Microsoft Copilot are routing enormous query volumes that would previously have ended at a search results page. For brands and content creators, the shift matters because your brand mentions is no longer defined by your rank position — it’s defined by whether the systems synthesizing answers trust you enough to cite you.
Ryan Jones, SVP of SEO at Razorfish, summed it up plainly on LinkedIn: “It’s not about the query anymore. It’s about semantic relevance to the topic.”
Nobody can agree on what to call this yet. GEO — generative engine optimization — was first named in a December 2023 academic research paper and has gained real traction in marketing circles. AEO, answer engine optimization, arrived shortly after and means roughly the same thing. Nikki Lam, VP and Head of SEO at NP Digital, coined the phrase “search everywhere optimization” in a widely-read 2024 article arguing that brands need to optimize across seven or more platforms beyond Google. The terminology is still shaking out. The underlying practice isn’t.
Creating passages for AI is the core of it. That means writing and structuring content so that large language models and AI-powered search systems can extract clean, accurate, attributable answers from what you publish. Not the page as a whole — specific passages. Discrete chunks of information that directly answer a question and can stand alone when pulled out of context.
Traditional SEO was deterministic. Put in the right inputs — keywords, backlinks, technical signals — get predictable outputs. Generative information retrieval is probabilistic. The system is pattern-matching against your brand’s overall topical authority, the clarity of your writing, the density of verifiable facts, and how often you’re mentioned and cited across the web as a whole. There’s no single lever to pull. Authority is built cumulatively, and it shows up in AI citations the same way it used to show up in PageRank — slowly, then all at once.
What does that mean practically? Every long-form piece of content you publish should include clearly labeled, information-dense sections that answer specific questions directly. Don’t build to your point. Make your point, then build around it. Declarative statements that can stand alone as answers. Consistent named entities — your brand, your authors, your product names — that allow AI systems to construct a coherent model of who you are and what you know. And original insight that doesn’t exist anywhere else in the training data, because that’s the only kind of content AI systems have an actual reason to surface over the generic stuff.
Lead with the claim. Substantiate after.
Google’s Martin Splitt offered an honest — if somewhat uncomfortable — explanation of the clicks situation in the AI Overview era: the clicks that do come through are increasingly likely to convert, because they come from more intentional, qualified users who’ve already been filtered by an AI response upstream. Useful framing. Doesn’t solve the attribution problem. Doesn’t help the journalism site, the niche publisher, or the local professional service business that needs traffic volume to survive.
The answer isn’t to stop doing SEO. It’s to stop treating it as the whole strategy.
Organic search (evolved SEO) still works. Technical health, high-quality content, authoritative backlinks, and strong E-E-A-T signals — Experience, Expertise, Authoritativeness, Trustworthiness — remain foundational. Google still processes roughly 5 trillion searches annually. Even if only 36% of those produce a click to the open web, that’s still an enormous number. The shift is in expectations and diversification, not in abandoning the channel.
AI search platforms are now a distinct distribution channel that requires deliberate strategy. ChatGPT, Perplexity, Gemini, and Microsoft Copilot all draw from the web to generate answers. Brands with strong representation in Wikipedia, high-authority industry publications, and consistent online mentions across credible sources are significantly more likely to appear in AI-generated responses. Your PR team and your SEO team are now, functionally, working on the same problem.
YouTube is the second-largest search engine in the world, and Google owns it. For brands that can produce quality video content, it offers a distribution channel that operates somewhat independently of organic search while remaining deeply integrated into Google’s broader ecosystem. Worth noting: AI systems are increasingly drawing on video transcripts as source material for responses, which means a well-captioned, information-dense video is also a passages for AI opportunity.
Community platforms — particularly Reddit — have become a meaningful search surface. Perplexity draws heavily from Reddit threads. Google’s own search results have elevated Reddit content significantly following recent algorithm updates. An active, genuinely helpful presence in relevant communities builds entity footprint in places that are now being indexed and cited by AI systems. This isn’t about creating a brand account and posting product announcements. Actual participation in conversations that matter to your audience.
Email and direct audience ownership. Full stop, this is the most algorithm-resistant distribution channel available. It bypasses search, bypasses social, bypasses AI intermediaries entirely. The inbox is yours. In a period of rising platform volatility, the direct relationship you’ve built with your subscribers is the asset least likely to get deprecated.
Podcasts. Transcripts are indexed. Guest appearances on well-known shows build authority signals that extend far beyond the listening audience — especially when those appearances are subsequently written about, linked to, or cited elsewhere.
Original data and proprietary research. If you publish a study nobody else has, journalists, bloggers, academics, and AI systems all have an incentive to reference you. This is probably the single highest-leverage investment in memorable influence available to any brand with the resources to do it. Proprietary research gets cited. Aggregated blog content doesn’t.
Good news buried in all of this: content that performs well for AI retrieval tends to be better content by every traditional measure too. Clearer. More direct. More useful. The practices that make your writing extractable by an AI language model are, almost without exception, the same practices that make human readers trust you.
A few specifics worth getting into.
Put your best information first. Not your “We-We” content. Not your background section. Not your “in today’s rapidly changing landscape” preamble. Your actual insight, your actual answer, your actual recommendation. AI systems extracting passages for AI favor content that front-loads key claims. So do people who’ve been reading online long enough to know that important information is often buried three paragraphs in.
Use descriptive, specific headings. “How to Optimize Your Content for AI Search Retrieval” does useful work. “More Tips” does none. The heading is a signal to both human readers and automated systems about what the section actually contains — treat it that way.
Author identity matters more than it used to. AI systems pick up on patterns. Content that is consistently attributed to named, credentialed people — with author bios, byline pages, and cross-references to other publications where that person’s expertise appears — builds a trust pattern that anonymous or byline-light content simply can’t. This is the practical, unglamorous work of E-E-A-T: making sure the humans behind your content are as visible and verifiable as the content itself.
Cite Gartner, Forrester, Pew Research Center, government data, peer-reviewed journals when they’re relevant to what you’re saying. Not because it impresses people, but because grounding claims in verifiable external sources is what separates content that AI systems cite from content that gets ignored. The reference list at the bottom of this article isn’t decoration.
And finally: topic clusters over isolated articles, every time. A single page on a subject doesn’t establish authority. A hub of interconnected, deep-coverage content — pillar pages linking to supporting articles, supporting articles linking back to pillars — creates a topical footprint that both traditional search algorithms and AI systems read as genuine domain expertise. One article is a contribution. Twenty well-linked articles on the same topic is a position.
Here’s something that doesn’t get enough attention in the average SEO conversation: search engines and AI systems don’t just process keywords. They build entity graphs — structured representations of who and what things are, how entities relate to each other, and how much trust the broader network assigns to those relationships.
Google’s Knowledge Graph, Wikidata, LinkedIn, Crunchbase, academic databases, industry publication bios, and citation networks all feed into how your brand is represented in these graphs. If your brand isn’t consistently named, described, and connected across those sources, you’re harder to place — and harder to cite.
Wikipedia, when warranted and when you meet their notability standards honestly, is still one of the most powerful entity signals available. NAP consistency — your Name, Address, and Phone number matching exactly across every business directory — matters particularly for local brands and any organization with a physical presence. Getting your subject-matter experts associated with specific topic areas through guest articles, conference appearances, and media interviews creates the kind of named-entity connections that AI systems use to assess topical authority.
Earn mentions, not just links. Links matter, they’ve always mattered. But unlinked brand mentions — your company name appearing in a trade publication, your CEO quoted in an industry newsletter, your product reviewed on a platform that doesn’t pass link equity — those are entity signals too. They contribute to the web of references that AI systems draw from when deciding whether your brand is a credible source on a given topic.
This is patient work. It doesn’t generate a traffic spike this Thursday. It builds the foundation that makes your memorable influence durable rather than fragile.
Attribution is genuinely broken for a growing share of the customer journey, and the honest thing to do is say so clearly rather than paper over it.
What Martin Splitt described is real: conversions are increasingly happening after an AI-generated response introduced a brand — with no search click in the chain connecting that impression to the outcome. A potential customer asks an AI about the best project management tools, sees your product cited in the response, does nothing immediately, Googles your brand name three days later, and converts on direct traffic. Your organic search report shows zero contribution. Your direct traffic report shows one conversion. The AI was the actual origin of the relationship and gets credited with nothing.
Traditional analytics models weren’t built for this.
The response isn’t to throw out traffic measurement — Cyrus Shepherd, founder of Zyppy SEO, made the point on LinkedIn with admirable directness: “clicks helped support creation, encouraged journalism, and ensured that no single entity controlled the flow of information.” Dismissing traffic as a vanity metric is a position that’s easiest to hold if you’re not a publisher, a journalist, or anyone whose income depends on people actually showing up to read what you wrote. Not every website is an e-commerce funnel. Some websites are how information gets distributed freely across the internet. That matters.
But the measurement stack needs to expand. Branded search volume. Direct traffic trends. Share of voice in industry publications. AI citation monitoring — tools that flag when your brand is mentioned in AI-generated responses are still early but improving fast. Social listening. Survey-based brand recall. These are the metrics that capture the parts of the customer journey now happening inside AI interfaces and word-of-mouth conversations, where a click was never part of the exchange to begin with.
John Mueller, one of Google’s longest-serving and most public search advocates, had an exchange with Martin Splitt about whether generative AI would eventually make having a website irrelevant. His answer was grounded in something easy to lose sight of when you’re deep in AI panic mode.
“Having a website is the basis for being visible in all of these systems,” Mueller said.
Chatbots need source material. AI Overviews pull from indexed web content. Local search results require a business presence online. E-commerce requires a place for transactions to actually happen. The underlying point — that a quality, authoritative web presence remains the foundation of visibility across every system, AI-powered or otherwise — is worth holding onto when the discourse gets apocalyptic.
What follows from that isn’t comfort, though. It’s a higher bar. Websites that get drawn from as source material, cited in AI responses, and trusted by search quality evaluators are the ones built for genuine usefulness — not the ones engineered to hit keyword density targets and accumulate links from content farms. The path forward is making your web presence so authoritative, so clearly structured, and so entity-rich that any system designed to find reliable answers finds you.
None of this is theoretical at this point. Here’s what the work actually looks like.
Audit your entity presence first. Search your brand name in Google, ChatGPT, Perplexity, and Gemini. Read what comes up. Is it accurate? Current? Are your key people associated with your brand’s actual areas of expertise, or does the information trail off into outdated bios and broken links? Identify what’s missing or wrong and fix the foundational stuff before investing heavily in new content.
Commission original research. A survey, a proprietary data analysis, an annual industry benchmark report. It doesn’t require a massive budget — it requires producing a finding that no one else has. Original data is the most citable content type available to any brand. It’s the thing journalists link to, the thing AI systems reference, the thing that earns your brand a footprint in topic areas where you were previously invisible.
Build topic clusters with intention, not with a content calendar. Every core subject your brand owns should have a pillar page supported by detailed, well-sourced articles that link back to it. A deliberate architecture that signals deep topical coverage — not a pile of loosely related posts that happen to use the same category tag.
Rewrite your existing content to lead with the claim. Go back to your ten most-trafficked pages and ask: does the most important information appear in the first two paragraphs? If the answer is no, fix it. This is the single fastest structural change that improves both human readability and AI extractability at the same time, and it costs nothing but editing time.
Start building your email list like it matters. Because it does. Choose one newsletter format that genuinely fits your audience, publish it consistently, and treat it as a direct line to the people who care most about what you’re building. Every subscriber is one fewer person whose attention you depend on an algorithm to reach.
Track what actually matters. Add branded search volume, direct traffic trends, share of voice, and AI mention monitoring to your regular reporting. The customer journey is no longer linear — your dashboards shouldn’t pretend otherwise.
Build for Citation, Not for Clicks. That’s the Whole Argument.
Michael King’s framing at SMX Advanced — SEO is deprecated, not dead — was precise enough to cut through two decades of “SEO is dead” noise. Deprecated software still runs. It still solves problems. It just isn’t where the platform is going, and betting the whole roadmap on it is increasingly a losing position.
The brands that build durable brand influence over the next decade are the ones that make themselves genuinely worth citing. Not worth ranking — worth citing. There’s a difference. Worth ranking is an optimization problem. Worth citing is a trust problem, a content quality problem, a brand reputation problem. It takes longer to solve and it compounds harder once you do.
Build content that actually helps people. A brand that real humans recommend in conversations. Distribution spread across enough channels that no single algorithm update can erase your visibility overnight. And treat everything you publish — every pillar page, every newsletter, every video, every podcast appearance — as another piece of evidence that you are the authoritative source on the topics you’ve decided to own.
The algorithm will keep changing. The audience won’t. Build for them.
Resources and further reading: SparkToro/Datos click-through rate researchGartner 2024 search traffic forecast; Google antitrust trial internal strategy memos (2024); SMX Advanced presentation by Michael King; Search Engine Land analysis by Danny Goodwin; Google’s John Mueller and Martin Splitt on AI, SEO, and the future of web visibility; LinkedIn commentary by Cyrus Shepherd (Zyppy SEO) and Ryan Jones (Razorfish); Nikki Lam, NP Digital, “Search Everywhere Optimization” (July 2024); Mustafa Suleyman, DeepMind co-founder, on conversational interfaces as the future of information retrieval; GEO research paper, December 2023.
