Google's AI Overviews are making a longstanding SEO problem more acute: a search result can deliver visibility without delivering a website visit. Pew Research Center found that users were substantially less likely to click links when an AI-generated summary appeared in Google results, raising a practical question for publishers and enterprise marketing teams. If a brand is surfaced in an answer but receives little measurable referral traffic, what should count as search success?
The answer is not to dismiss AI Overview visibility. It is to separate answer presence from audience acquisition. Search teams that continue to rely primarily on rankings, impressions and organic clicks risk treating distinct outcomes as if they were interchangeable. Google's evolving results pages require more deliberate measurement, clearer ownership and content governance that accounts for AI-mediated discovery.
What Pew's findings say about AI Overviews and clicks
In its March 2025 analysis of Google search behavior, Pew found that AI Overviews appeared in about 18% of searches. When an AI summary was present, about 8% of searches led to a click on a traditional search result, while roughly 1% led to a click on a link within the AI summary. Across Google searches overall, Pew reported that roughly two-thirds ended without a click.
The figures show correlation in observed search behavior, not a universal rule for every query, region or business. Still, the direction is clear: the presence of an AI-generated answer is associated with lower click-through to the wider web. The underlying methodology and findings are detailed in Pew Research Center's analysis of Google users and AI summaries.
| Search outcome | Observed rate | Context |
|---|---|---|
| Searches showing an AI Overview | About 18% | Pew's March 2025 search analysis |
| Click on a traditional result when an AI Overview appeared | About 8% | Queries with an AI-generated summary |
| Click on a link within the AI Overview | About 1% | Queries with an AI-generated summary |
| Google searches ending without a click | Roughly two-thirds | Overall finding in Pew's analysis |
Industry discussion often places zero-click behavior in a broader range of roughly 60% to 70% for general search, with some analyses reporting higher rates for AI Overview queries. Those ranges should not be treated as a single benchmark. Results can vary by geography, query type and changes Google makes to AI Overviews and related experiences such as AI Mode.
For SEO leaders, the important change is less about one headline percentage than the changing relationship between being found, being cited and being visited. A result page can satisfy an informational need before a user reaches a publisher's site. That may support awareness, but it may not produce the sessions, leads or conversions that conventional organic reporting is designed to track.
A governance problem, not just a traffic problem
AI Overview visibility creates a measurement challenge because a traditional click is a clear event, while appearing as part of an AI-generated answer may be difficult to connect to downstream commercial value. Counting every appearance as a win would overstate performance. Ignoring those appearances would miss a growing layer of search exposure.
A more useful governance framework distinguishes three questions:
- Visibility: Does the brand, product or source material appear in relevant AI-assisted search results?
- Engagement: Does that appearance lead to clicks, branded searches, return visits or other observable interactions?
- Business value: Can the search program be connected to qualified demand, leads, revenue or another agreed outcome?
This approach helps prevent an impression metric from replacing a business metric. It also makes ownership clearer. SEO, content, analytics, product marketing and legal or compliance teams may all have a role when corporate information is being summarized in AI-driven results.
Content strategy should similarly move beyond publishing for ranking alone. Organizations need accurate, well-structured material that directly addresses important customer questions and clearly identifies the entity, offering and evidence behind key claims. That does not guarantee inclusion in an AI Overview, and Pew's data does not establish a formula for earning citations. It does, however, reinforce the value of content that can serve both a human reader and a search system seeking a concise, reliable answer.
Measurement needs to become more explicit about its limits. Teams can monitor organic traffic and conversions alongside search visibility, branded demand and the pages that continue to attract high-intent visitors. They should also document where AI Overview exposure is observed and avoid presenting it as equivalent to a referral visit. Reporting that separates these signals is more useful for budget decisions than a single blended KPI.
For businesses whose growth depends on informational search, the immediate task is to identify which queries are most exposed to answer-first results and which still create meaningful click opportunities. The next task is to decide where authority-building content, conversion-focused pages and measurement investment can have the greatest commercial impact.
For enterprises, AI-generated search answers can weaken the link between content investment and easily visible referral traffic. Scalevise helps teams assess whether their brand is represented across AI answer experiences, identify gaps in high-value topics and connect visibility work to practical governance. Use the AI Visibility and GEO Checker to establish a clearer baseline before revising your search KPIs. Start an AI Visibility scan.
Frequently Asked Questions
Do Google AI Overviews reduce clicks to websites?
Pew found that searches showing an AI Overview were associated with low click-through rates: about 8% clicked a traditional result and about 1% clicked a link within the AI summary in its March 2025 analysis.
How often did AI Overviews appear in Pew's analysis?
AI Overviews appeared in about 18% of the Google searches examined by Pew in March 2025.
Should an AI Overview appearance count as SEO success?
It can indicate visibility, but it should not be treated as equivalent to a website visit, lead or conversion. Teams should measure visibility, engagement and business value separately.
Is there one universal zero-click rate for AI search?
No. Reported rates vary by query type, geography and changes to Google's search experiences. The broader figures provide context, not a universal benchmark.
Conclusion
Pew's findings show that AI Overviews can reduce the likelihood of users clicking through from Google to traditional web results. For enterprise SEO, that makes click-based reporting incomplete rather than irrelevant. The stronger response is to govern AI search visibility as a distinct signal, evaluate its relationship to engagement and commercial outcomes, and keep content strategy grounded in accurate, useful information.
