How AI Is Changing Organic Search (And What To Do About It)
Search results don’t look like they used to. Google now answers many questions directly on the results page, generated by AI, before a single blue link appears. For a lot of businesses, that means fewer clicks even when rankings haven’t moved.
If you’ve noticed organic traffic dipping while your rankings stay steady, you’re not imagining it. This is happening across almost every industry we work in, from finance to retail to B2B services. The good news is that this shift is manageable. It just requires a different kind of visibility, not just a different kind of ranking.
There’s a second, quieter problem underneath this. As AI tools answer questions without a click, standard analytics is missing more of the picture too, which is worth keeping in mind as you read the traffic numbers below.
Quick answer: AI Overviews are cutting organic click-through rates by 34.5% to 58% for top-ranking results (Ahrefs), with the steepest drops on informational searches, while transactional searches remain largely unaffected. The fix is to restructure content so it can be cited inside AI answers (clear structure, structured data, genuine E-E-A-T), back it with technical foundations like llms.txt, and make sure your measurement can still see what’s working.
What’s Actually Happening to Organic Traffic
Google’s AI Overviews (and similar features in Bing) summarise an answer directly in the search results, pulling from multiple sources at once. The person often gets what they need without clicking through to any website.
Ahrefs analysed Search Console data across 300,000 keywords and found that click-through rates for the number one organic result dropped by 34.5% when an AI Overview appeared on the page. A follow-up analysis found that drop had widened to 58% by the end of 2025. The steepest drops sit on informational searches (the “what is”, “how to”, “best way to” type questions). Transactional searches, where someone is ready to buy or book, are far less affected. That’s an important distinction, and it changes how we think about content strategy.
The other shift is where research happens at all. More people are asking questions directly inside ChatGPT, Perplexity, and Google’s own AI tools rather than typing into a search bar. Being visible in a traditional search result is no longer the only game. Being cited as a source inside an AI-generated answer is now just as valuable, sometimes more so, because it carries an implicit endorsement.
This doesn’t mean organic search is dying. It means the definition of “ranking well” is expanding to include showing up inside AI answers, not just above them.
How Brands Can Appear Inside AI Answers
AI tools build their answers from content they trust and can easily understand. That gives us a genuine playbook, not guesswork. Here’s what we’re prioritising for clients right now.
- Structured, citation-worthy content
AI systems favour content that answers a question clearly and early, in plain language, backed by a credible source. Long, meandering intros before the actual answer get skipped over. Clear headings, direct answers in the first sentence or two, and well-organised data all improve the odds of being pulled into an AI response.
- Technical foundations (structured data and llms.txt)
Schema markup helps AI tools and search engines understand exactly what your content is and who it’s from. It’s not new, but it matters more now than ever.
There’s also growing use of an llms.txt file, a simple text file that tells AI crawlers which parts of your site are the most authoritative and worth referencing. It’s early days for this standard, but for brands with a lot of content, it’s a low-cost way to guide AI tools towards your best material rather than your thinnest pages.
- E-E-A-T signals, done properly
Google has talked about Experience, Expertise, Authoritativeness and Trust for years, but AI Overviews have made it non-negotiable. Real author bios, genuine case studies, original data, and clear business credentials all help AI systems decide your content is worth citing. Generic, unattributed content rarely makes the cut.
- Digital PR and mentions across the web
AI tools cross-reference how often a brand is mentioned across the internet, not just what’s on your own site. Coverage in reputable publications, industry directories, and third-party reviews all build the kind of trust signal that gets a brand cited inside an AI answer.
- Blog content optimisation for direct answers
We’re now writing (and re-writing existing) blog content so that each section can stand alone as a complete answer to a specific question. That means clear H2s phrased as questions, a direct answer immediately underneath, then supporting detail. This structure works for human readers and AI summarisation equally well. This sits within our broader content marketing approach, which treats every piece as both a ranking asset and an AI citation opportunity.
Why This Matters More Than Your Ad Spend
Here’s the part worth sitting with. For a lot of businesses, agency and content fees make up a larger share of the marketing budget than paid media does. Organic search and content are a long-term asset. If AI is changing how that asset gets discovered, protecting it isn’t optional, it’s where the actual return on your investment lives.
One practical piece of advice worth flagging here: as AI answers pull people away from clicking through, and browsers keep restricting cookies and referrer data, standard analytics is seeing less of the picture. Moving to server-side tracking gives you an accurate read on how organic, paid and AI-driven traffic are actually performing, rather than judging results off a shrinking, less reliable sample. For businesses spending real money on SEO and content, that’s the difference between decisions made on real numbers and decisions made on guesswork.
The businesses that get ahead here won’t be the ones panicking about traffic dips. They’ll be the ones who treat this as what it is: a content quality and structure problem with a clear set of fixes, not an unsolvable algorithm mystery.
Key Takeaways
- AI Overviews are reducing click-through rates on informational searches, but transactional searches are largely unaffected
- Being cited inside an AI answer is becoming as valuable as ranking highly in traditional results
- Structured data, llms.txt, and genuine E-E-A-T signals all improve your odds of being referenced
- Content should be restructured to answer questions directly and early, not buried in a long introduction
- Digital PR and third-party mentions strengthen how AI tools perceive your credibility
- As a value-add, moving to server-side tracking gives you an accurate read on performance as AI answers and privacy changes erode standard analytics data
- This is a solvable, strategic problem, not a reason to abandon organic search
Where We Come In
This is exactly the kind of shift we help clients navigate, because it sits across SEO, content, and technical strategy all at once. We’re already building AI visibility audits into the work we do for clients like yours.
If you want a clear picture of how your site currently shows up (or doesn’t) in AI-generated answers, get in touch with our team. We’ll walk you through exactly where the opportunity is.
