Google AI Overviews Are Stealing Your Clicks

Google AI Overviews have fundamentally changed what it means to rank. A page that sits in position one no longer guarantees meaningful traffic if an AI-generated summary above it answers the query before anyone scrolls. The practical consequence is that organic click-through rates are falling for informational searches, and marketers who built their SEO strategy around traditional ranking signals are now seeing the results in their analytics.

This is not a future risk. It is a present one. The question worth asking is not whether AI Overviews will affect your traffic, but how much they already have, and what a rational response looks like.

What Are Google AI Overviews and How Do They Work?

AI Overviews are the AI-generated summaries that appear at the top of Google search results pages for eligible queries. Google generates them by synthesising information from multiple sources across the web, presenting a consolidated answer directly in the search interface. The sources used are sometimes cited with links, but the summary itself is designed to resolve the query without requiring a click.

They were launched broadly in the United States in May 2024 under the name AI Overviews, having previously been tested as Search Generative Experience. Since then, Google has expanded their availability across more countries and query types, though the rollout has been uneven and the product continues to evolve. Some queries trigger an Overview consistently. Others never do. The logic is not fully transparent, which creates its own planning challenge.

From a technical standpoint, the system draws on Google’s large language models combined with its existing web index. It is not simply scraping the top-ranked pages. The synthesis process can pull from sources that do not rank in the top ten for a given query, which means your position in traditional results does not guarantee inclusion, and inclusion does not require a strong traditional ranking. That asymmetry is important.

If you want to understand how AI is reshaping search and content strategy more broadly, the AI Marketing hub at The Marketing Juice covers the commercial implications across the full stack, from content production to paid media to measurement.

Which Types of Content Are Most Affected?

Not all queries are treated equally. AI Overviews appear most frequently on informational searches: how-to questions, definitions, comparisons, and factual lookups. These are precisely the query types that content marketing programmes have targeted for years because they generate high search volume and map neatly to top-of-funnel awareness objectives.

The irony is not lost on me. When I was growing the SEO function at a previous agency, we built entire content calendars around informational queries because they were accessible, measurable in terms of rankings, and easy to justify to clients. The strategy worked well for a long time. What it was really doing, in retrospect, was capturing demand that existed rather than building any durable competitive advantage. AI Overviews expose that fragility directly.

Transactional queries, where someone is ready to buy or take a specific action, are less frequently accompanied by AI Overviews. The same applies to highly localised searches, queries involving real-time information, and searches where Google’s system determines that a synthesised answer would not serve the user well. This gives commercial-intent content a degree of protection that informational content no longer has.

Brand and navigational queries are largely unaffected. If someone searches for your brand name directly, they are not looking for a synthesised answer. They want to reach you. That traffic remains relatively stable.

The content categories sitting in the most difficult position are those that answered common questions well but offered nothing beyond the answer itself. Glossary pages, FAQ content, and generic how-to articles are the clearest examples. If the content was created primarily to rank for a query rather than to serve a specific audience with genuine depth, AI Overviews will replace it efficiently.

What Does the Traffic Impact Actually Look Like?

The honest answer is that it varies considerably by site, sector, and query mix. Anyone claiming a precise universal figure is working from a sample that may not apply to your situation. What the available evidence suggests, including analysis from tools like Ahrefs’ AI and SEO research, is that click-through rates on queries that trigger AI Overviews are lower than equivalent queries without them. The degree of suppression depends on where your result sits, how prominently the Overview is displayed, and whether your site is cited within it.

Being cited inside an AI Overview does drive some traffic. It is not zero. But the volume is typically lower than what position one generates without an Overview present, and the citation pattern is inconsistent enough that you cannot build a reliable content strategy around chasing it.

This is where measurement discipline becomes critical. I have spent a significant part of my career trying to get marketers to measure what actually matters rather than what is easy to report. The risk right now is that teams see a decline in organic sessions, attribute it to AI Overviews, and restructure their entire content programme in response, when the actual cause might be a broader shift in search behaviour, a technical issue, or seasonal variation. Before changing strategy, you need to understand which queries are triggering Overviews for your specific site, whether your impressions have held while clicks have fallen (the clearest signal of Overview suppression), and how your competitors are faring on the same queries.

Google Search Console is the starting point. Filter by query type, compare click-through rates over time, and look for the specific pattern of stable or growing impressions alongside declining clicks. That is the fingerprint of Overview suppression. Without that analysis, you are making strategic decisions based on assumption rather than evidence, which is a habit the industry has never fully shaken.

Can You Optimise to Appear Inside AI Overviews?

Yes, to a degree, though the approach is less mechanical than traditional SEO optimisation and the results are less predictable. Google has not published a definitive set of criteria for Overview citation, but the patterns that emerge from observation align with what you would expect from a system designed to surface trustworthy, well-structured information.

Structured content performs better. Clear headings, concise answers to specific questions, and logical information hierarchy all make it easier for Google’s systems to extract and attribute content. This is not new advice. It is the same logic that drives featured snippet optimisation, and much of that groundwork transfers directly.

Authority and trust signals matter significantly. Sites with strong E-E-A-T signals, demonstrable expertise, clear authorship, and a track record of accurate information are more likely to be cited. The Moz research on AI content points to the growing importance of these signals in an environment where AI-generated content is proliferating and Google needs reliable ways to distinguish authoritative sources from noise.

Original data and genuine expertise are harder to synthesise than general information. If your content contains proprietary research, original analysis, or perspectives that cannot be assembled from existing sources, it is more likely to be cited and less likely to be replaced entirely. This is the direction content investment needs to move in. Generic informational content produced at scale to capture search volume is increasingly a commodity that AI can generate itself. Content that reflects real expertise, specific experience, and original thinking is not.

There is also a competitive intelligence angle worth considering. Moz’s work on LLM competitive research and gap analysis makes the case that understanding how AI systems perceive your brand and your competitors is becoming a distinct research discipline, separate from traditional keyword analysis. Knowing which queries trigger Overviews that cite your competitors but not you is actionable information.

How Should SEO Strategy Change in Response?

The strategic response is not to abandon SEO. It is to be more precise about what you are trying to achieve and more honest about which content investments are defensible.

I have seen this pattern before in different contexts. When I was at lastminute.com, we ran paid search campaigns that generated six-figure revenue within a single day from relatively simple executions. The mechanics were straightforward because the intent signals were clear and the path to conversion was short. The lesson was not that paid search was magic. It was that matching the right message to a clearly defined intent, at the right moment, produces results. The same logic applies to SEO in an AI Overview world. Content that serves a specific, well-defined intent with genuine depth will outperform content that chases volume without clarity of purpose.

Bottom-of-funnel content deserves more investment. Product pages, comparison content, case studies, and content that sits close to a purchase or conversion decision are less exposed to Overview suppression and more directly connected to revenue. If your content programme has been heavily weighted toward top-of-funnel informational content, the economics of that investment have changed.

Content that builds audience rather than just capturing search traffic becomes more valuable. Email lists, direct relationships, and content that people actively seek out are not subject to algorithm changes. I have watched agencies lose significant client revenue because the entire traffic strategy was built on a single channel. Organic search was that channel for many businesses. The diversification argument was always valid. It is now urgent.

For teams thinking about how AI tools fit into their content production process, resources like Semrush’s guide to AI content strategy provide a practical framework for integrating AI assistance without losing the editorial quality that makes content worth citing in the first place. The goal is not to produce more content faster. It is to produce content that earns authority.

Technical SEO fundamentals still matter. Schema markup, clean site architecture, fast load times, and clear entity signals all contribute to how Google understands and trusts your content. These are not new priorities, but they are more important when the system evaluating your content is making synthesis decisions rather than simply ranking pages.

What About the Broader AI Search Landscape?

Google AI Overviews are the most immediate concern for most marketers because Google retains the dominant share of search traffic. But the broader shift toward AI-mediated search is happening across multiple platforms simultaneously. Perplexity, ChatGPT’s search functionality, and Microsoft Copilot in Bing are all routing queries through AI synthesis rather than traditional link lists. The pattern is consistent even if the market share varies.

This matters for how you think about content strategy at a structural level. The question is no longer just “how do we rank?” It is “how do we ensure our brand and our expertise are represented accurately when AI systems synthesise answers in our category?” That is a different question, and it requires a different kind of thinking.

Brand mentions, citations in credible publications, and a consistent presence in authoritative sources all influence how AI systems represent your business. This is sometimes called generative engine optimisation or GEO, though the terminology is still settling. The practical implication is that traditional PR, digital PR, and content distribution into third-party publications are now SEO-adjacent activities in a way they were not before.

For teams exploring AI tools more broadly, Ahrefs’ AI tools webinar series covers the practical application of AI in SEO workflows, including how to use these tools without creating the kind of undifferentiated content that AI Overviews will simply absorb and replace.

One thing I have learned from judging the Effie Awards is that the campaigns that hold up under scrutiny are the ones with a clear connection between strategy and outcome. The same discipline applies here. If you cannot articulate why a specific piece of content will drive a specific business outcome, the fact that it might rank or get cited in an AI Overview is not a sufficient justification. The measurement framework has to come first, not as an afterthought.

What Should You Measure and How?

The measurement challenge with AI Overviews is real. Google Search Console shows you impressions and clicks but does not explicitly flag which queries triggered an Overview. You have to infer it from the pattern. Third-party tools are beginning to fill this gap, with platforms like Semrush and Ahrefs building features that track Overview presence for specific queries, but the data is still maturing.

The metrics worth tracking are click-through rate by query type, impression volume versus click volume over time, and ranking position alongside Overview presence. If your impressions are holding but your CTR has dropped significantly on a cluster of informational queries, that is strong evidence of Overview suppression rather than a ranking problem. The response to those two situations is completely different, and conflating them leads to misallocated effort.

I have spent a lot of time in my career arguing that most marketing measurement is more comfortable than accurate. Teams report the metrics that look good rather than the ones that reveal what is actually happening. The AI Overview impact is a case where the temptation to avoid looking closely is high, because what you find might require uncomfortable changes to content programmes that took years to build. Look anyway. The businesses that understand their actual position will adapt. The ones that avoid the analysis will keep investing in content that is generating diminishing returns.

There is a wider conversation about how AI is changing the measurement and attribution landscape across marketing channels. The AI Marketing section of The Marketing Juice covers these developments as they evolve, including the implications for how marketing teams report performance and justify investment.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions