Ahrefs published updated CTR data in February 2026 showing that pages ranking #1 in Google now get 58% fewer clicks when an AI Overview appears above them. That single stat should reframe how you think about SEO for the rest of the year.
The impact of AI search on SEO isn’t theoretical anymore. It’s showing up in traffic dashboards, conversion data, and revenue reports across every industry. We studied over 500 high-value digital marketing topics and tracked how AI search platforms: Google AI Overviews, Google AI Mode, ChatGPT, Claude, and Perplexity, are shifting where clicks go, which pages get cited, and how much those visitors are actually worth.
Some of what we found confirmed what most SEOs suspected. And some of it contradicts the advice the industry has been repeating for years.

Traffic from AI search platforms is on track to surpass traditional organic search traffic for digital marketing topics by early 2028, based on our projections. And that timeline could accelerate fast if Google makes AI Mode the default experience, which they’re already testing.
The adoption numbers tell the story. ChatGPT’s weekly active user base grew 8x between October 2023 and April 2025, hitting over 800 million. Google’s share dipped below 90% for the first time since 2015.
But combined traffic (AI + traditional) won’t just stay flat. It’ll probably decline first, then stabilize.
Why? AI search compresses the funnel. Users get answers without clicking anything. And links inside AI responses are less prominent than blue links in traditional results. seoClarity’s research found that AI Overviews hit 30% of U.S. desktop keywords by September 2025, a 492% year-over-year jump. That’s a huge chunk of queries where the traditional click path is being rerouted or eliminated.
I’ve watched this play out with clients who saw their informational content traffic drop 20-30% on queries where AI Overviews appeared, while their conversion-focused pages barely moved. The impact isn’t uniform. It depends on what kind of content you’re publishing and who you’re targeting.
Start tracking your visibility inside LLMs now. Not next quarter. Platforms like AI-focused SEO tools can show you where your brand appears (or doesn’t) across ChatGPT, Perplexity, and Google’s AI features.
Traditional SEO and LLM optimization overlap, but they aren’t identical. The biggest gap? Traditional SEO rewards full pages. AI systems reward extractable chunks of information. If you haven’t restructured your content for answer engine optimization, you’re already behind.

An AI search visitor converts at 4.4x the rate of a traditional organic visitor, based on tracked non-Google AI sources like ChatGPT.
That number surprised me. I expected higher engagement, not a conversion rate multiplier that dramatic.
The logic makes sense once you think about it, though. By the time someone clicks through from an AI response, they’ve already done their comparison shopping inside the chat. The LLM summarized their options, maybe even explained your pricing or value proposition. They’re arriving at your site with a decision mostly made.
AI responses also feel like personal recommendations. A Pew Research study from July 2025 found that Google searchers who saw AI Overviews were substantially less likely to click on regular results links. But the people who did click through were more purposeful; they’d already been pre-qualified by the AI response.
We project that by the end of 2027, AI search channels will drive roughly the same total economic value as traditional organic search globally. And it’ll keep growing from there.
Even small traffic gains from AI search can move the needle on revenue because these visitors are closer to buying. The priority isn’t volume. It’s getting cited.
That means making your content easy for LLMs to quote and reference. Present pricing, specs, and value props in clean, structured formats. Keep brand mentions consistent across every channel where AI systems pull training data. Publish machine-readable data through schema markup and structured pages that search engines can interpret.

Not as much as you’d think. And this is where the data gets interesting.
When ChatGPT cites webpages, about 90% of the time those pages rank in traditional search positions 21 or lower. They’re not even on page one of Google. Perplexity and Google’s own LLMs also frequently cite lower-ranking pages.
But there’s a twist: some LLMs cite pages in positions 1-5 more often than pages in positions 6-20. So ranking at the top still helps, but there’s a massive gap in the middle where traditional rankings don’t seem to matter much for AI citation.
We have three theories for why lower-ranked pages get cited so frequently.
The simplest: there’s just more content to choose from below position 20. It’s a bigger pool. The more significant reason is that AI systems are designed to surface discrete pieces of information, not whole pages. A perfectly structured FAQ answer buried on a page ranking #47 can beat a thin top-3 result because the AI can extract exactly what it needs.
seoClarity’s 500-million-keyword study did find that over 99% of Google AI Overview citations come from pages in the top 10 organic results. So Google’s own AI features still lean heavily on traditional ranking signals. ChatGPT and Perplexity? Much less so.
Traditional SEO isn’t dying; Google search volume actually grew 21%+ even as AI features expanded. But the goal is shifting. You need to optimize for citation, not just ranking.
That means creating content for specific, niche use cases. Product guides for narrow verticals. Blog posts answering the oddly specific questions your support team hears every week. The more targeted your content, the more likely an AI system is to pick it for a response that matches a particular user’s situation. Consider how GEO and SEO overlap rather than treating them as separate strategies.

Quora. That’s the #1 most-cited domain in Google AI Overviews, according to our data. Reddit comes in second.
Both platforms are goldmines for the exact type of content AI systems love: real questions from real people getting specific, detailed answers that aren’t covered elsewhere. Nobody writes a 2,000-word blog post answering “Is it worth replacing a water heater myself if I have no plumbing experience but watch a lot of YouTube?” But someone on Reddit does.
Reddit’s performance likely gets an extra boost from Google’s data partnership, which gives Google direct access to Reddit content for training.
The rest of the top 20 most-cited domains are high-authority sites, names like NerdWallet, Travel + Leisure, and Good Housekeeping. These aren’t small publishers. They’re domains with deep E-E-A-T authority and decades of content.
If you’re ignoring Quora and Reddit as marketing channels, you’re leaving AI visibility on the table. Participating in relevant threads (authentically, not spamming) puts your brand in front of the content AI systems are already pulling from.
Getting mentioned on other high-authority domains matters too. Digital PR and quality backlink strategies aren’t just for traditional rankings anymore. Brand citations on cited domains increase the chance that AI systems mention you by name when generating responses about your category.

Our research found that 50% of links in ChatGPT 4o responses point directly to business and service websites. Even though forums like Quora and Reddit dominate per-domain citation counts, the aggregate link distribution favors commercial sites.
This finding held across other models we tracked.
The takeaway: LLMs heavily reference your own website when responding to queries about your business or industry. They view business sites as legitimate topical sources, not just commercial pages to be skeptical of.
Your website is your most powerful AI visibility asset. But only if it’s built for how AI systems read content.
Focus on content quality and originality. LLMs value the same things search engines value, just through different mechanisms. Create multimodal content; text, images, video, audio, so AI systems have multiple ways to interpret and surface your pages.
Structure content for natural language processing. That means clear headings, named entities, simple grammar, and topic-relevant internal linking. Many AI crawlers won’t execute JavaScript, so make sure your pages are crawlable without it. Understanding AEO versus GEO strategies helps you pick the right optimization tactics for each AI platform.
And publish comparison guides. LLMs frequently generate comparison-style responses, and they’ll pull heavily from pages that already organize information that way.
| AI Search Factor | Traditional SEO Impact | AI/LLM Citation Impact |
| Page ranking position | Primary success metric | Matters for Google AIOs, less for ChatGPT/Perplexity |
| Content structure & extractability | Moderate (helps UX) | High, direct-answer blocks get cited 4-5x more |
| E-E-A-T signals | Important for ranking | Heavily weighted for citation selection |
| Schema markup & structured data | Helps rich snippets | Provides machine-readable claims AI can verify |
| Niche, specific content | Good for long-tail traffic | Preferred by LLMs over generic broad content |
| Brand mentions across web | Builds authority signals | Directly increases AI brand recognition |
AI search is reshaping traffic patterns, click behavior, and how users make buying decisions. The data makes that clear. But the correction I’d make to most of the panic-driven commentary: SEO isn’t dying. It’s splitting.
Traditional ranking optimization still matters for Google AI Overviews, which pull 99% of citations from top-10 results. But ChatGPT, Perplexity, and other LLMs are building their own citation logic, one that rewards content specificity, structured formatting, and genuine brand authority over raw domain power.
The businesses that will win this transition aren’t the ones chasing every new AI feature. They’re the ones building content that works for both systems, pages that rank traditionally and get extracted by AI. That’s where the 4.4x conversion multiplier lives. And it’s where the future of SEO is heading, whether you’re ready for it or not.
Does AI Search Actually Reduce Organic Traffic?
Yes, but the effect varies by query type. Ahrefs’ December 2025 data showed a 58% CTR drop for pages ranking #1 when an AI Overview appeared above them. Informational content takes the biggest hit. Conversion-focused and transactional pages have been relatively stable in our tracking, because users still need to visit the site to take action.
How Do I Get My Pages Cited in Google AI Overviews?
Rank in the top 10 first. seoClarity’s study of 500+ million keywords found that over 99% of AI Overview citations come from pages in positions 1-10. Beyond ranking, use hierarchical heading structure, include attributed statistics, and add schema markup. Google’s AI features pull from the same index as traditional search, so strong SEO fundamentals are your entry ticket.
Is AI Search Traffic More Valuable Than Traditional Organic Traffic?
Based on conversion rate data, AI search visitors are worth 4.4x more than traditional organic visitors. They arrive with more information and clearer purchase intent because the LLM already walked them through their options. Even modest AI traffic gains can outperform larger volumes of traditional organic visits in terms of revenue impact.
Do Backlinks Still Matter for AI Search Visibility?
They do. High-authority domains dominate the top-cited lists in both Google AI Overviews and ChatGPT responses. Backlinks remain a core authority signal that influences which sources AI systems trust. But brand citations, mentions of your business name without a hyperlink, are gaining weight in AI systems. A brand mention on a cited forum thread can influence AI responses even without a direct link.
Why Is Quora Cited More Than Major Publications in AI Overviews?
Quora and Reddit answer highly specific, niche questions that traditional publishers rarely cover. AI systems are designed to match specific user intent, and these platforms provide granular answers for long-tail queries. Reddit also benefits from Google’s data partnership, which gives Google direct training access to Reddit content.
Should I Optimize Separately for ChatGPT and Google AI Overviews?
Yes. Our data shows these systems cite differently. Google AI Overviews pull almost exclusively from top-10 ranked pages. ChatGPT cites pages from position 21+ about 90% of the time. A page that doesn’t rank well in Google can still get heavy ChatGPT citation if its content is well-structured and answers a specific question. Covering both requires different content strategies for each system.
Is SEO Dead Because of AI Search?
No. Google search volume grew over 21% even as AI features expanded. SEO is evolving, not disappearing. The shift is from optimizing only for rankings to optimizing for both rankings and AI citations. Businesses that treat this as an expansion of their SEO strategy, rather than a replacement, are seeing the best results in 2026.

Michael Vale has over 5 years of experience helping clients improve their business visibility on Google. He combines his love for teaching with his entrepreneurial spirit to develop innovative marketing strategies. Inspired by the big AI wave of 2023, Michael Vale now focuses on staying updated with the latest AI tools and techniques. He is committed to using these advancements to deliver great results for his clients, keeping them ahead in the competitive online market.