You have improved your website, published useful content, and followed search engine optimization (SEO) best practices. Yet your organic search traffic keeps slipping. Before deciding that your strategy—or your marketing team—has failed, consider another possibility: the search results page itself has changed.
Independent research shows that people are less likely to click traditional results when Google's AI Overviews appear. That creates a frustrating possibility for business owners: a page can remain visible in search while earning fewer visits.
There is meaningful evidence behind that explanation. There is not enough evidence to assume that AI causes every traffic decline, or that Google's recent stock-market losses directly measure the clicks businesses have lost. Here is what the research supports, what Alphabet's earnings actually tell us, and how to investigate your own website.
What AI search changes about the click
Google distinguishes between AI Overviews, which summarize a topic within search results, and AI Mode, which supports more conversational exploration and complex questions. Both can include links to supporting websites, but they are different experiences. Findings about one should not automatically be treated as measurements of the other. Source: Google Search Central
For a business, the practical concern is straightforward: if a search result answers someone's immediate question, visiting the source website becomes optional. Google itself acknowledges that some quick-answer searches can be satisfied without another click. Source: Google, August 6, 2025
Imagine a potential customer researching what a technical term means. A clear summary may be enough. Someone evaluating a service provider, investigating a complicated problem, or looking for original evidence may still need the website. That distinction is useful when deciding which lost visits matter most to your business.
Alphabet's selloff is relevant, but it does not prove the whole theory
The major post-earnings selloff occurred on Thursday, October 29, 2026, not Friday, October 30. CNBC reported a 13% decline and roughly $500 billion erased from Alphabet's market capitalization. Zacks reported a 12.64% decline and $482 billion in lost market value. Those reports differ on the market-value estimate; neither number represents advertising revenue lost. Sources: CNBC and Zacks
There is a genuine AI-search connection. During Alphabet's October 28, 2026 earnings call, chief business officer Philipp Schindler said growing AI Mode usage affected Search growth because AI Mode had lower monetization. He also said Google was expanding ads within AI Overviews and AI Mode. But Schindler identified foreign exchange as the single largest driver of Search deceleration from Q2 to Q3, alongside retail and media-and-entertainment advertising dynamics. Source: Alphabet's Q3 2026 earnings call
Search advertising revenue did not collapse: CNBC reported that it grew 12% year over year to $63.35 billion, although it missed analyst expectations. CNBC also reported negative quarterly free cash flow of $9.07 billion and full-year capital expenditure guidance between $195 billion and $205 billion. The market reaction involved spending, cash generation, and growth expectations—not simply fewer link clicks. Source: CNBC
The defensible connection is that AI is changing both search behavior and Google's monetization economics. The evidence does not establish that AI Overviews taking clicks from Google Ads was the main cause of the selloff.
It is also essential to separate paid ad clicks from unpaid organic visits. Evidence that an AI summary reduces clicks to traditional website results does not, by itself, quantify Google's lost advertising revenue.
The clearest evidence: people click less when AI summaries appear
Pew Research Center measured actual browsing behavior
Pew's July 22, 2025 analysis used browsing data from 900 U.S. adults during March 1–31, 2025. Its dataset contained 68,879 unique Google searches, with 12,593 producing an AI summary in the results Pew collected. Source: Pew Research Center
The key findings were:
- Traditional search-result links received clicks on 8% of visits when an AI summary appeared, compared with 15% without one.
- Links inside the AI summary received clicks on 1% of visits to pages with a summary.
- Browsing sessions ended after 26% of pages with AI summaries, compared with 16% of pages containing only traditional results.
These are different measures, not percentages to add together. In particular, ending a browsing session is not the same as every possible kind of no-click search. Source: Pew Research Center
Query type also mattered. AI summaries appeared for 8% of one- or two-word searches, 53% of searches with 10 words or more, 60% of searches beginning with question words, and 36% of searches containing both a noun and a verb. Those findings make question-answering content a sensible place to start investigating exposure—not a reason to assume every long search loses traffic. Source: Pew Research Center
Ahrefs found lower click-through rates even for top-ranking pages
In a study published February 4, 2026, Ahrefs analyzed 300,000 keywords: 150,000 with an AI Overview and 150,000 informational keywords without one. It compared aggregated desktop Google Search Console click-through rates from December 2023 and December 2025. Click-through rate, or CTR, measures clicks relative to impressions. Source: Ahrefs
Ahrefs found that an AI Overview's presence correlated with a 58% lower average CTR for the top-ranking page. Its estimate used changes in the comparison group to model the expected CTR without AI Overviews; it was not a finding that every website lost 58% of its total traffic. Source: Ahrefs
The reported CTR impacts by ranking position were:
- Position 1: -58.0%; position 2: -50.8%.
- Position 3: -46.4%; position 4: -38.8%.
- Position 5: -32.6%; position 6: -30.5%.
- Position 7: -29.7%; position 8: -28.8%.
- Position 9: -29.7%; position 10: -19.4%.
For business owners, the important distinction is between winning a ranking and receiving the click. Those are related outcomes, but they are not interchangeable.
What these studies cannot tell you
Neither study was a randomized experiment assigning identical searches to results with and without AI. Pew observed March browsing behavior but reconstructed search results during April 7–17, 2025; it explicitly notes that summaries can change. Ahrefs compared desktop keyword groups across December 2023 and December 2025, using a modeled baseline. These limitations matter when interpreting cause and applying averages to an individual business. Sources: Pew and Ahrefs
Google offers a different aggregate perspective. In its August 6, 2025 statement, the company said total organic clicks to websites were relatively stable year over year and that click quality had increased. Google defined quality clicks as visits where people did not quickly return to search. It also acknowledged that traffic was shifting between websites. That dated, company-reported statement is not proof that your leads or revenue improved. Source: Google
Aggregate stability and losses for particular pages can coexist. The studies support taking AI-related click loss seriously; they do not establish a universal decline of the same size across all websites.
How to diagnose your own traffic decline
Start with Google's Search Console data rather than assuming that a falling analytics graph explains itself. Google's troubleshooting guidance recommends reviewing the last 16 months, comparing the last 3 months with the previous period and year over year, and examining affected queries, URLs, countries, devices, and search types. Source: Google Search Central
Use this investigation sequence:
- Locate the loss. Separate informational articles from service, product, and branded searches. Identify which pages and queries account for the change rather than relying only on a sitewide average.
- Compare visibility with clicks. If comparable queries retain impressions and rankings but lose CTR, inspect their actual results pages. AI answers are one hypothesis; competing snippets and other search features are alternatives. Stable visibility alone does not prove AI caused the loss.
- Check demand and technical health. Review Google Trends, indexing, crawling, security issues, manual actions, and recent URL migrations. Google lists algorithm changes, technical problems, security and spam issues, seasonality, changing interests, and site moves as possible causes.
- Inspect AI visibility separately. Google's June 3, 2026 announcement introduced generative AI performance reports and says they reached all websites worldwide on August 31, 2026. The announced fields are impressions, pages, countries, devices for Search results, and dates with hourly, daily, weekly, and monthly views. Use them to understand exposure, not to calculate lost clicks from impressions alone.
- Connect traffic to business outcomes. Compare qualified inquiries, purchases, and other meaningful conversions alongside visits. Google recommends complementing Search Console with conversion and engagement tracking in analytics tools. Fewer visits and fewer customers are not automatically the same problem.
Sources: Google's traffic troubleshooting guide, generative AI report announcement, and AI features documentation
Treat these checks as a diagnosis, not a shortcut to blaming AI. Multiple causes can occur together.
What to change, and what not to abandon
Keep the SEO fundamentals. Google says there are no additional technical requirements or special optimizations needed to appear in AI Overviews or AI Mode. Pages still need to be indexed and eligible for a search snippet. Google recommends crawlable pages, internal links, accessible textual content, a good page experience, accurate structured data, and current Business Profile and Merchant Center information where relevant. It says no special AI text file or schema.org markup is required. Eligibility does not guarantee inclusion. Source: Google Search Central
Give visitors a reason to go beyond the summary. As an editorial recommendation, prioritize original examples, documented experience, useful tools, detailed comparisons, and clear explanations of your actual services. A generic definition and a detailed buying decision serve different needs. Google also identifies original posts, in-depth reviews, and firsthand perspectives as content people seek out. None of these formats guarantees a citation or a click. Source: Google's traffic perspective
Measure the result you actually need. Keep traffic in the dashboard, but place it beside qualified leads, sales, and conversion performance. Do not automatically celebrate a higher conversion rate if the total number of valuable inquiries is falling. Likewise, do not scrap useful content solely because low-intent visits declined.
Reduce dependence on a single source of discovery. As a business-risk recommendation, consider how email, customer referrals, partnerships, and direct repeat visits fit your audience. This is diversification, not a promise that another channel will replace lost search traffic.
Good SEO still matters. The click is no longer guaranteed.
There is credible evidence that Google's AI answers are associated with lower click-through rates. That means declining traffic can reflect a changed search experience, not simply bad content or neglected SEO.
But explanation is not diagnosis. Before rewriting your website or replacing your strategy, establish whether the loss comes from rankings, demand, technical issues, reduced click-through, or a combination. Then focus improvements on helping the right people make decisions and become customers.
If your reports show fewer visits despite continued SEO work, bring the page-level evidence to your website and marketing team. The useful question is not just, “Why aren't we ranking?” It is, “When we are visible, what still gives someone a reason to visit, and what happens after they do?”
Research reviewed November 2, 2026. Study dates are stated above because search interfaces and user behavior continue to change.
AI Disclosure: This article was developed with assistance from AI tools and reviewed by our team prior to publication.