October DealsAmazon USOctober deal check: compare before you payAmazon US: current deals, useful picks and tech finds.Check DealsWindows FixRecommendedWindows errors stealing your time? Find the fix fastScan stability, cleanup and performance issues.Fix NowOctober DealsAmazon USDeal season is back - check today's better picksAmazon US: current deals, useful picks and tech finds.See Picks×
Skip to content
The Finance Base
The Money Desk · Blog
Re:

Google Says AI Search Hasn’t Cut Overall Website Clicks. Publishers Still Face Losses.

Google says organic clicks from Search remain broadly stable, but aggregate totals can hide losses at individual publishers. The key questions are who gets the clicks, what users do after they arrive, and how much those visits earn.
From TheFinanceBase Team7 min to read
Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Google says the total number of organic clicks Search sends to websites is “relatively stable” year over year, but that does not mean every publisher is unharmed. In an August 6, 2025 statement, Google also acknowledged that traffic is shifting between sites, with some losing visitors while others gain them. The dispute is therefore less about whether AI affects traffic at all than about which sites lose clicks, and whether more engaged visits can make up for fewer of them.

What Google actually denied

Google Search chief Liz Reid said on August 6, 2025, that aggregate organic click volume from Google Search to websites was “relatively stable” year over year. Google also said the average quality of those clicks had increased slightly, defining a quality click as one after which a user does not quickly return to Search. The statement did not publish the underlying data or quantify the claimed improvement. Google’s statement also said it sends billions of clicks to websites every day; that is Google’s claim, not a contemporaneous independent audit.

Crucially, Google did not say all publishers’ traffic was stable. It acknowledged that traffic is moving between sites, with some gaining and others losing. A stable total can coexist with significant losses for particular publishers, sectors, or types of search query.

AI Overviews and AI Mode are not the same feature

AI Overviews

AI Overviews are AI-generated summaries shown within ordinary Google Search results, alongside links to web pages. They can answer a question before a user visits a site. By May 20, 2025, Google said the feature was available in more than 200 countries and territories and more than 40 languages; that is a dated availability statement, not a guarantee of current availability in every location. Google’s May 2025 update described that expansion.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

AI Mode

AI Mode is a more conversational Search experience intended for longer, multi-part, follow-up, and multimodal questions. Google said it uses “query fan-out”: issuing multiple related searches across subtopics before assembling a response. Google announced its U.S. rollout without a Labs sign-up on May 20, 2025; interface and availability can change. Google’s AI Mode announcement also said AI Overviews had driven more than a 10% increase in usage for query types that show them in the U.S. and India. That was Google’s reported comparison for those qualifying queries, not a measure of all searches.

Why Google says AI search can help the web

Google’s case is that summaries may encourage people to ask more, longer, or more complex questions; AI results can surface additional links; and visitors who click after seeing a summary may be more motivated or spend longer on the destination site. Query expansion could also create opportunities for pages to appear for questions users might not previously have asked. Google made a related argument in August 2024, saying links in AI Overviews produced higher-quality clicks and that an experiment placing links directly in the AI response increased traffic to publisher sites. Google’s 2024 account is evidence of the company’s position, not proof that the effect is universal.

Google also argues that some reports of traffic decline rely on isolated examples, flawed methods, or changes that began before AI features appeared. That is a relevant challenge to causal claims, but without disclosed underlying data it does not settle the issue either.

Why stable clicks do not settle the publisher question

Totals can hide who is losing

If one group of sites gains visits while another loses them, the total can remain steady. Google’s own acknowledgment that traffic is shifting makes the distribution of clicks central: independent publishers, newsrooms, affiliate sites, forums, retailers, and large platforms may experience very different outcomes.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Click quality is not click volume

A more engaged visit may be more valuable for a subscription, lead, or purchase business. But an ad-supported publisher often depends on total visits, page views, impressions, and repeat browsing. One engaged landing-page visit may not replace several visits or pages that an on-page answer has made unnecessary. Google did not disclose the size of the quality improvement, and “quality” does not automatically mean higher revenue.

Clicks are not revenue

To understand the business impact, separate search impressions and clicks from landing-page sessions, pages per session, engaged time, newsletter sign-ups, purchases, leads, advertising impressions, affiliate conversions, and subscription starts. Search Console clicks and analytics sessions are also different measurements. Even stable clicks would not establish stable revenue if visitors browse less, convert less often, or arrive on pages with different monetization.

A zero-click search does not necessarily mean an AI answer

A zero-click search is one where a user does not click through to an external website from the results page. The user may have seen an AI Overview, but may instead have used a knowledge panel, featured snippet, weather result, sports score, map, calculator, or another direct answer—or simply left without clicking. Google compared AI answers with older answer features such as the Knowledge Graph and sports scores. To assess AI’s contribution, zero-click rates need to be separated by result type, query, device, geography, and date.

What the reported zero-click evidence does—and does not—show

TechCrunch reported that Similarweb data showed the share of news searches ending without a click to a news website rose from 56% in May 2024 to 69% in May 2025. TechCrunch’s report is a useful signal of a changing search experience, but the figures do not prove that AI Overviews alone caused the increase or quantify the effect across the entire web. Changes in search behavior and the news market also matter.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Likewise, a sitewide traffic decline cannot by itself prove that AI caused it. Ranking changes, seasonality, demand, news cycles, social-platform algorithms, direct visits moving into apps, Google Discover changes, technical performance, and measurement changes can all affect a publisher’s numbers. The reverse is also true: stable aggregate clicks cannot prove that no publisher has been harmed.

Which kinds of pages face the greatest exposure?

More exposed: answers that fit in a summary

Pages are more vulnerable when their main value is a short, generic answer that can be reproduced directly in Search: definitions, basic factual explanations, simple how-to content, generic listicles, and some basic health or consumer explainers. Weather, dates, sports scores, and calculations are other direct-answer cases. Google cited “when is the next full moon” as a question a user may resolve without clicking onward.

Potentially more resilient: distinctive value beyond the answer

Google says users are seeking forums, videos, podcasts, first-hand perspectives, in-depth reviews, original posts, and first-person analysis. Original reporting, proprietary data, expert interpretation, tested comparisons, community discussion, interactive tools, and transactional pages can also give a visitor a reason to go beyond a summary. These are not guarantees of traffic protection: they are types of value an answer box may not fully deliver.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

How publishers can measure the effect on their own site

Start with Google Search Console for page- and query-level impressions, clicks, click-through rate, country, and device data. Then connect those changes to sessions and business outcomes in an analytics system. Google Search Console is useful for diagnosing Search visibility, but it cannot by itself establish revenue impact or prove AI causation.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
  1. Build comparable periods. Compare equivalent time windows and account for seasonality, major news cycles, and changes in demand.
  2. Segment query groups. Separate queries where AI Overviews appear from those where they do not, and distinguish informational, navigational, and transactional intent. Google Search Console does not provide a complete, definitive record of AI Overview exposure for every impression, so treat any classification accordingly.
  3. Split branded from non-branded searches. Growth in branded searches can mask losses in generic discovery.
  4. Break out device, country, and language. Search features and user behavior can differ across these dimensions.
  5. Separate channels and surfaces. Do not mix Google Search with Discover, News, Images, YouTube, social referrals, or direct traffic when judging Search changes.
  6. Follow the path to revenue. Compare clicks and sessions with engagement, pages viewed, sign-ups, leads, purchases, advertising yield, and affiliate conversions.
  7. Compare page types. Look for differences between commodity explainers and pages with original reporting, first-hand experience, tools, or transactional intent.

Common analytical traps include treating a lower click-through rate as a revenue loss without conversion data, using averages that conceal the typical publisher’s experience, and equating longer time on site with greater commercial value. A before-and-after comparison can help identify where to investigate; without a suitable comparison group or other controls, it does not isolate AI as the cause.

What the dispute means for publishers and marketers

The economic trade-off depends on the business model. Fewer but more intentional visits might work for a publisher whose revenue depends on qualified leads, purchases, or subscriptions. A business built on broad discovery, advertising impressions, or page depth may be hurt if a summary replaces the browsing journey, even if the remaining visitors are more engaged. That is a plausible implication of Google’s definition of quality clicks and its admission that traffic is shifting—not a proven outcome for every site.

For finance publishers, the distinction is especially practical: a brief explanation of a widely answered money question may be easier to summarize than a tested product comparison, original analysis, or tool that helps a reader make a decision. The relevant question is not simply whether a page ranks, but whether a Search visit still leads to a valuable reader relationship or business outcome.

Reducing dependence on a single referral source is also a business decision, not an AI-specific SEO trick. Publishers can assess whether useful search visitors become newsletter subscribers, members, customers, or repeat direct users. Any audience or subscription platform should be chosen for the publisher’s scale, commerce needs, deliverability, migration needs, and payment geography—not as a universal fix.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.

Leave a Reply

Your email address will not be published. Required fields are marked *

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

More post from the Money Desk

  1. The Money DeskBlogTheFinanceBase09 OCT 267 minMortgage Escrow FAQs: Taxes, Insurance, Shortages, and Refunds
  2. The Money DeskBlogTheFinanceBase09 OCT 265 minHow Mortgage Escrow Accounts Work and What Homeowners Pay For
  3. The Money DeskBlogTheFinanceBase09 OCT 265 minHow to Read a Stock Chart, Volume and Market-Cap Data
Recommended PC Tool
Recommended PC Tool
Windows Errors? Fix Them Before They SpreadFree repair scan
Crashes, No Sound, or Screen Glitches?Free driver scan

Two free Windows tools

One Free Minute Could Fix That PC

Before you go - each of these free tools takes about a minute and tackles what quietly slows a Windows PC down.

Special offer. View Outbyte info, uninstall instructions, EULA, and Privacy Policy.