SEO Isn't Dying. It's Turning Into Organic Search

Mateusz Rzetecki

For years, SEO came down to one question: where do you rank on Google? Today that question is losing its meaning. The click we fought for is disappearing from the results page — and with it, the whole logic of "ranking."

The numbers leave little doubt. According to an Ahrefs study from December 2025 (300,000 keywords, Google Search Console data), the presence of an AI Overview cuts the top-ranking page's CTR by 58% — up from 34.5% in early 2025. A Pew Research study (68,879 searches from 900 U.S. adults, March 2025) found that when an AI answer appears, only 8% of users click a traditional result, versus 15% without one. And per Similarweb, roughly 60% of Google searches now end with no click at all.

This isn't the "death of SEO." It's a new playing field. Search isn't going away — what's changing is where and how people find answers. That's why SEO in the narrow sense is becoming organic search: building consistent brand presence everywhere AI can learn about you.

What the data says

An AI Overview cuts the position-one CTR by 58% (Ahrefs, December 2025).
When an AI answer appears, only 8% click a traditional result, down from 15% (Pew Research, March 2025).
Brand mentions correlate with AI visibility far more than links: 0.664 vs. 0.218 (Ahrefs, 75,000 brands).
Brands in the bottom half of web mentions are essentially invisible to AI (Ahrefs).

From ranking on Google to presence everywhere

Classic SEO had one channel and one metric: your Google rank. Organic search is a different model. A language model doesn't "index" your site the way Google's crawler did — it learns your brand from all the text circulating about you online. From forums, from video transcripts, from industry articles, from comparison sites, from discussions.

In practice, that means moving from "how do I rank higher?" to "how is AI supposed to know I exist — and who should it recommend me to?" In a generative model there is no "position 3." There's frequency: how often your brand shows up in answers to hundreds of relevant prompts. You're either there or you're not.

Where AI actually learns about your brand

The strongest evidence came from Ahrefs' study of 75,000 brands. The factor most strongly correlated with presence in AI Overviews was brand mentions across the web (0.664) — more than three times stronger than traditional backlinks (0.218). In the December 2025 update, YouTube mentions took the lead (correlation around 0.737) — the single strongest signal of all. The reason is mundane: both Google and OpenAI trained their models on YouTube transcripts (per the "New York Times," GPT-4 learned from over a million hours of footage).

The second key source is user-generated content. In a Semrush analysis (30 million sources, reported by Search Engine Land), Reddit was the most-cited source across ChatGPT, Google AI Mode, Gemini, Perplexity and AI Overviews — with YouTube, LinkedIn, Wikipedia and Forbes also near the top. No accident: in 2024 Google signed a data-licensing deal with Reddit estimated at around $60 million a year.

The third takeaway is about who writes about you. Muck Rack's analysis of over a million AI-cited links found that 82% came from earned media — independent, editorial content, not your own pages or paid placements. The single most important sentence from all this research: brands in the bottom half of web mentions are essentially invisible to AI — no matter how good their SEO is.

One brand, many surfaces

If AI learns from the whole web, your own site isn't enough. You need consistent presence across every surface the model draws knowledge from. In my experience, it pays to map them as one plan, not as separate, disconnected efforts:

Your own blog and site — your only fully controlled source of truth. This is where you define the facts AI should repeat.
YouTube — today's most strongly correlated signal. Your brand name in the title, description and transcript measurably raises your odds of being cited.
Reddit and industry communities (Quora, Stack Exchange, niche groups) — the "credibility layer" AI treats as the voice of real users.
Earned media — editorial coverage, expert commentary, original data and studies others want to cite. This carries 82% of AI presence.
Comparison and review platforms (G2, reviews, industry rankings) — where AI checks how you stack up against competitors.
Social media and LinkedIn — repeated, consistent messaging reinforces recognition of your brand as an entity.

The goal isn't to "be everywhere" for its own sake. The goal is for your brand to be described the same way in every one of those places.

Brand attributes first, content second

This is the step most often skipped — and in my view it decides everything. Before you publish anything, define your brand attributes: what you are, for whom, in what category, what makes you different, the words you want to be described with. A short, hard set of facts, for example:

category: "a tool for one-time AI-search visibility reports,"
for whom: brands and marketing teams checking their presence in ChatGPT, Gemini, Perplexity and AI Overviews,
differentiator: a one-time report instead of a costly ongoing subscription,
3–5 phrases you use to describe yourself consistently.

Why does this matter so much? A language model builds its picture of your brand from the co-occurrence of words and from the consensus of many sources. If you call yourself one thing on your blog, another on LinkedIn and a third in an interview, you blur the entity. AI doesn't know who you are, so it won't recommend you with conviction. But when the same consistent narrative repeats everywhere, the model gets the same fact confirmed many times over — and starts treating it as the truth about the category. Consistency here is, quite literally, a ranking signal.

The practical rule I follow: one "brand brief" on a single page, identical for PR, content, social and anyone who speaks publicly. Every asset reinforces the same attributes instead of reinventing the brand.

How to start — and how to measure it

You can't improve what you don't measure. The sequence I recommend is simple:

Write your brand attributes on one page and align them across every channel.
Run a presence audit: which prompts mention you, how AI describes you, where competitors win.
Pick the 1–2 surfaces with the biggest gap (usually YouTube and earned media) and concentrate your effort there.
Track mention frequency over time — that's your new metric instead of rank.

Step two is the hardest today, because models are non-deterministic — the same question yields five different answers. That's why at aisearchinsight.io we put together a one-time AI-search visibility report: it shows how ChatGPT, Gemini, Perplexity and AI Overviews describe your brand right now, which prompts you appear in, and where competitors are ahead. A snapshot of "where you stand today" that's a sensible place to start.

To be clear: we're not an agency and we're not selling you SEO or GEO services. Our job is to give you an honest picture of the situation. If the report shows action is needed, we have trusted partners who can run that work — but the decision and the direction stay with you.

After more than a decade in this field, I've seen plenty of "revolutions." This one is different, because it changes the unit of visibility itself: from the link to the mention, from the position to the presence. The brands that win will treat their visibility as one coherent system — not a collection of separate channels.

Sources

Mateusz Rzetecki

Brings over 16 years of experience in crafting and executing innovative SEO and content marketing strategies to enhance brand visibility and drive organic growth across diverse industries.

WWW LinkedIn