Google's AI Overviews have been live in the UK and US for a while now, and the panic in the SEO world has been real. Traffic drops, zero-click searches, publishers losing referral visits - it has all been blamed, at least in part, on these AI-generated answer boxes that now sit at the top of many search results pages. The question a lot of business owners and marketers are asking is a fair one: if Google just answers the question itself, what happens to the websites below it?
The short answer is: traditional blue-link results are not going away. But they are being pushed down, filtered, and in some cases replaced entirely for specific types of queries. The more useful question is not whether traditional results will disappear, but which searches are changing, how fast, and what you can actually do about it.
How AI Overviews actually work
Google AI Overviews are generated using a large language model that pulls from indexed web content to produce a synthesised answer at the top of the results page. Unlike a featured snippet, which lifts a chunk of text from a single page, an Overview draws from multiple sources and rewrites the information in its own words. Sources are cited below the generated text, but most users never click them.
Google has been running variations of this for years. The Knowledge Graph, featured snippets, and "People Also Ask" boxes all reduced clicks before AI Overviews arrived. What is different now is the scale and the sophistication. AI Overviews appear for a much wider range of queries, they are longer, and they give users enough information to act without visiting any external site at all.
From a technical standpoint, Google's AI Overviews pull primarily from pages that are already well-ranked. The model is not scraping random corners of the web. It is reading the same content that ranks on page one and synthesising it. That creates a strange dynamic where ranking well on Google can actually feed the system that reduces your clicks.
The types of queries most affected
Not every search is treated the same. AI Overviews are far more common for certain categories of query than others.
Informational queries
These are hit hardest. "How do I clean a cast iron pan?", "What is the difference between a sole trader and a limited company?", "Why is my sourdough too dense?" - for questions like these, an AI Overview can answer the user entirely. The page that used to rank for those queries gets cited, maybe, but clicked far less often. Studies published in 2024 and early 2025 have shown click-through rate drops of between 15% and 64% for pages appearing beneath an AI Overview, depending on the query type and position.
Navigational and transactional queries
These are much less affected, at least for now. When someone searches "buy running shoes size 10 UK" or "FlinnSchema pricing", they have clear intent. Google knows an AI paragraph is not what they need. Traditional product listings, ads, and brand pages still dominate these results. That is not going to change soon because Google's ad revenue depends on it.
Local queries
Searches with local intent ("best Italian restaurant in Leeds", "emergency plumber near me") still produce the local pack and map listings prominently. AI Overviews do appear in some local searches, but they tend to sit above the map pack rather than replacing it. The bigger risk for local businesses is not Google AI Overviews - it is ChatGPT and Perplexity answering those queries directly, without any map pack at all.
What the data is actually showing
It is worth being honest about the data here, because a lot of commentary has been either alarmist or dismissive. The reality sits somewhere in the middle.
Google's own data, as presented at various industry events, shows that AI Overviews are increasing overall search satisfaction and session length. Users ask follow-up questions, explore more, and reportedly click through more often when an Overview sparks curiosity rather than closing a topic down.
Independent data tells a more complicated story. Analyses from firms like BrightEdge and studies published in Search Engine Land have found that organic click-through rates for position one results have declined meaningfully in categories where AI Overviews are common. Some publishers, particularly those producing high-volume informational content, have reported traffic drops of 20% to 40% year-on-year.
The honest summary: AI Overviews are reducing clicks for informational content. They are not significantly affecting transactional or navigational search yet. And they are still rolling out, so the picture will keep shifting.
Traditional results are not going anywhere, but their role is changing
Here is the thing that often gets lost in the conversation: Google still shows traditional results on almost every page that includes an AI Overview. The links are there. The snippets are there. They are just lower on the page, and users' attention is absorbed by the Overview before they reach them.
The role of traditional results is evolving from being the primary answer to being a next step for users who want to go deeper. That is a meaningful shift. It means content that is genuinely detailed, opinionated, or experience-based will still earn clicks, because the AI Overview can point to it without fully replacing it. Thin informational content, on the other hand - the 600-word "what is X?" articles that used to rank easily - is increasingly redundant.
Google has also been explicit that it wants to send traffic to publishers. Whether it follows through on that commitment as AI Overviews expand is a fair sceptical question, but the current design does include citations and does generate some referral clicks, just fewer than before.
The bigger picture: AI search engines beyond Google
Focusing only on Google AI Overviews misses a larger shift happening in parallel. ChatGPT, Perplexity, Gemini, and other AI tools are being used directly to answer the kinds of questions that used to go to Google. A user who asks ChatGPT "what is the best project management software for a small agency?" never even visits a search results page. No click-through. No organic traffic. No chance for your page to appear.
This is where the conversation about AI Overviews has to expand. Google is changing, yes. But the more disruptive trend is that search itself is fragmenting. People are doing searches that never touch Google at all. And for that audience, traditional SEO provides almost no visibility.
That is the problem we spend most of our time on at FlinnSchema. Getting businesses visible inside AI tools - not just on Google's results page, but in the answers that ChatGPT, Perplexity, and Gemini give directly to users. Understanding what AI visibility actually means is a useful starting point if you have not thought about this distinction before.
What you can do about it
Sitting back and waiting to see what happens is a valid choice if your traffic is primarily transactional and local. But for most businesses, there are things worth doing now.
Audit which of your pages are losing traffic to AI Overviews
Google Search Console will show you pages with declining impressions and click-through rates. Cross-reference those with queries that now trigger an AI Overview (you can check this manually by searching your target terms). If a page is being answered by an Overview, you have a decision to make about whether to fight for that query or focus effort elsewhere.
Make your content harder to fully summarise
First-hand experience, specific case studies, proprietary data, strong opinions, and detailed how-to content are all harder to replace with a generated Overview. An AI can summarise the general principle of how sourdough fermentation works. It cannot replicate your bakery's 14 years of trial and error. That kind of content still earns clicks because users want more of it.
Add structured data so AI systems can understand your content precisely
Schema markup does not directly influence whether you appear in an AI Overview, but it does help AI systems - Google's and others - correctly interpret what your page is about. Well-structured pages with clear entity relationships, FAQ schema, HowTo schema, and product schema are more likely to be cited accurately, and more likely to be pulled into AI answers across platforms.
Choosing the right types of schema for your business is not complicated once you understand the categories, and it pays dividends across Google, ChatGPT, and Perplexity simultaneously.
Build visibility in AI tools beyond Google
If your potential customers are using ChatGPT or Perplexity to find products and services in your category, you need to be visible there. That means structured data, clear business entity signals, consistent information across platforms, and content that AI models can actually parse and attribute to your brand. A free AI visibility audit will tell you where you currently stand across the main AI search tools, not just Google.
Rethink your content strategy for queries that are now zero-click
If a query is now reliably answered by an AI Overview and your traffic data confirms you are not getting clicks, continuing to produce thin informational articles for that query is a waste of resource. Redirect that effort toward transactional content, comparison content, case study content, and community-driven content that AI cannot easily generate on its own.
The bottom line on AI Overviews
Traditional search results are not being replaced. They are being demoted and contextualised differently. For transactional searches, the impact is minimal right now. For informational searches, the impact is real and growing. For searches that skip Google entirely and go straight to an AI tool, traditional SEO offers no protection at all.
The businesses that will handle this shift well are those treating it as a prompt to build deeper, more distinct content and to extend their visibility into AI tools beyond Google. The ones that will struggle are those waiting for the dust to settle before doing anything differently.
If you want to understand how your business is currently performing across AI search tools, there are practical ways to check for yourself before investing in any formal audit or service.
Frequently Asked Questions
Will Google AI Overviews kill organic search traffic?
Not entirely, but they are reducing clicks for informational queries. Transactional and navigational searches are much less affected. The more pressing threat to overall organic traffic is AI tools like ChatGPT and Perplexity handling searches that never reach Google at all.
Does appearing in a Google AI Overview help or hurt my site?
It is complicated. Being cited in an Overview gives your brand exposure, but may reduce direct clicks to your page. Studies show that citation positions do drive some traffic, but far less than a traditional rank one result would. The main benefit is authority association - users see your brand mentioned alongside other reputable sources.
Can I opt out of being included in Google AI Overviews?
Not directly. You can use robots.txt directives or meta tags to limit how Google crawls and indexes your content, but this would also remove you from traditional search results. There is currently no way to appear in traditional results while being excluded from AI Overviews. Most businesses are better served by optimising for both rather than attempting exclusion.
How is this different from the older featured snippets problem?
Featured snippets lifted a specific passage from a specific page, and the source was highly visible. AI Overviews synthesise content from multiple sources, rewrite it in new language, and cite sources below the fold. The click-through impact is generally worse than featured snippets because the answer is more complete and the source attribution is less prominent.