🔍 Read the full analysis: Does ChatGPT Really Account For A Fifth Of Google’s Searches? on ThorstenMeyerAI.com
Get smart everyday buys delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
OpenAI consumer product head Adam Fry said ChatGPT’s search-like query volume is roughly one-fifth of Google’s, but did not identify the Google measure, time period or countries behind the comparison. Independent models estimate lower ratios, while website traffic panels show Google sending far more visitors than ChatGPT. The figures measure different things and do not independently verify OpenAI’s claim.
OpenAI consumer product head Adam Fry said Nov. 2 that ChatGPT’s volume of search-like queries is roughly one-fifth of Google’s query volume, a comparison that has not been independently verified because OpenAI did not disclose the Google figure or its calculation. Independent estimates cited by ThorstenMeyerAI.com put ChatGPT’s search-like activity below that ratio, while website traffic data shows a much larger gap between Google and ChatGPT referrals.
Fry made the claim during a fireside chat with journalist Alex Kantrowitz at an OpenAI-hosted dinner in New York. Big Technology published the conversation on Nov. 3. Fry described the measure as “search-like queries,” rather than saying that ChatGPT and Google receive an identical kind of search request. He explained that a question counts when a user asks it and ChatGPT searches the web.
The definition matters. According to Search Engine Journal and Search Engine Roundtable, Fry said: “Every time you ask a question and go search the web, that’s a search.” The Verge reported that simple requests such as defining a word or asking for a basic how-to explanation do not count under that description. Fry did not specify which Google query total OpenAI used, what period or countries were covered, or provide underlying data. Google did not comment on the claim in the reports cited by the source.
Two independent estimates described in the source are lower, though both are models, not direct measurements. Ahrefs estimated about 12% in February 2026, using a judgment-based estimate of how many ChatGPT prompts resembled Google searches. SeoClarity estimated about 7.5% on Oct. 28, 2026, by modeling ChatGPT messages and comparing them with an estimate of Google’s annual searches. The methodologies and definitions differ, so neither figure is a like-for-like audit of OpenAI’s number.
Infographic 01 · ChatGPT as a share of Google
Questions or clicks? ChatGPT is a fifth of Google, or less than 2 percent
OpenAI says one fifth. Independent models say less. Traffic panels say far less, because a chat answer often ends without a visit.
Counting questionsHow many search-like questions?
Counting clicksHow many visits reach websites?
All values: ChatGPT as a percentage of Google on the same yardstick. OpenAI’s figure is a company claim (striped bar); the two models are estimates; the panels see only their own sites. Bars scaled to 20 %.
Sources: Big Technology, 3 Nov 2026 · Ahrefs, 11 Feb 2026 · seoClarity, 28 Oct 2026 · Ahrefs traffic panel, Oct 2026 · StatCounter, Oct 2025–Oct 2026.
Powered by Thorsten Meyer AI
Search Queries Versus Website Visits
The one-fifth comparison can sound like ChatGPT has taken a fifth of Google’s business, but it does not establish that. It compares a stated volume of search-like activity with an unspecified Google query figure. It says nothing on its own about advertising revenue, search-market share, or how many users choose ChatGPT instead of Google.
The distinction is particularly relevant to publishers and businesses that depend on visitors arriving from search engines. In an Ahrefs panel covering 118,671 websites in October 2026, Google accounted for 24.74% of visits and ChatGPT for 0.36%. That is a measure of traffic reaching sites in the panel, not a count of queries made inside either service. StatCounter’s comparison of referrals to sites in its network likewise showed ChatGPT at 0.08% to 0.41% of Google-versus-ChatGPT clicks in each month from October 2025 through October 2026.
Those panels suggest that ChatGPT-generated visits were small relative to Google referrals in the measured samples. They do not prove how many queries either service handles: users can receive an answer without clicking a source, and the panels cover participating websites rather than every site. For companies, the practical reading is to track AI visibility and search referrals separately, rather than treating one as a replacement for the other.
Products Worth Considering
Dell UltraSharp 32 U3226QER 4K Thunderbolt Hub Monitor
Canon PIXMA PRO-200S Wireless Professional Inkjet Photo Printer
FLEXISPOT ErgoX Ergonomic Office Chair with Footrest, Black
As an affiliate, we earn on qualifying purchases.
How the Competing Estimates Differ
The source places OpenAI’s statement alongside several measures that answer different questions. OpenAI’s ratio concerns queries that prompt a web search within ChatGPT, according to Fry’s explanation. Ahrefs’ February estimate began with a figure of 2.5 billion ChatGPT prompts per day from an OpenAI usage study, then applied its own estimate that 65% of those prompts were things people might search for on Google. Ahrefs’ Patrick Stox characterized that percentage as his judgment and said it could be lower depending on the definition.
SeoClarity’s Oct. 28 estimate used a different approach: it extrapolated ChatGPT message volume using user growth, counted all “seeking information” messages and a quarter of “practical guidance” messages as search, then compared that total with an estimate of Google’s searches. SeoClarity described the result as “a modeled comparison, not a measured market-share figure.” These methodological choices help explain why the independent percentages do not line up precisely with one another or with OpenAI’s claim.
OpenAI’s published usage data offers another reference point, but not a direct check of the one-fifth ratio. The source says 20.2% of consumer ChatGPT messages in September 2026 were categorized as “seeking information.” That category is not necessarily the same as web searches: some information requests may not trigger browsing, while other categories could include questions that do. The source’s rough arithmetic says about six in ten total messages would need to trigger a web search for the one-fifth comparison to hold under its assumptions. That is an interpretation, not an OpenAI-published measurement.
Google’s own recent performance also complicates claims of a simple handoff from search to AI chat. Alphabet reported that Google query growth in the third quarter was its fastest in five years, and search revenue rose 12% to $63.35 billion. These company-reported results do not resolve the comparison with ChatGPT, but they show Google’s search business was still growing during the period described.
““A modeled comparison, not a measured market-share figure.””
— SeoClarity
The Missing Google Denominator
The central unknown is how OpenAI arrived at the one-fifth ratio. Fry did not identify the Google query figure, measurement period, geographic coverage or underlying dataset, and the source reports no independent access to OpenAI’s counts. Google did not comment in the reports cited. The statement itself is confirmed as something Fry said; its numerical accuracy is not independently confirmed by the available material.
There is also no single agreed definition of a “search” across the comparisons. A prompt may resemble a search without triggering ChatGPT’s web tool, and a web lookup may answer a question without sending a user to an external site. Conversely, a traffic panel records visits, not every question asked. The models and panels therefore cannot be combined into one definitive market-share figure. The source also does not provide enough methodological detail to reproduce its rough calculation based on OpenAI’s September message categories.
What Would Verify the Ratio
A stronger check would require OpenAI to publish the count of ChatGPT questions that actually trigger web searches, the period and countries covered, and the precise Google query total used for comparison. Independent researchers would then need access to comparable definitions and data. Until that happens, the one-fifth figure should be treated as OpenAI’s stated comparison, not an independently established measure of search-market share.
For readers and businesses, the available next step is to keep the measures distinct: follow query estimates as estimates, and use referral analytics to see whether AI tools are sending visitors to particular sites. The figures in this report are tied to the dates and samples provided; later usage patterns or broader panels could produce different results. When an AI answer cites sources and the underlying information matters, readers can check those sources directly rather than assuming the answer reflects a completed visit to the original site.
Key Questions
Did OpenAI confirm that ChatGPT handles one-fifth as many searches as Google?
Adam Fry made the claim at an OpenAI-hosted event on Nov. 2, 2026. The source material does not independently verify the ratio, and OpenAI did not disclose the Google figure or calculation used.
What counts as a ChatGPT search in Fry’s explanation?
Fry said a search is counted when a user asks a question and ChatGPT searches the web. The Verge reported that simple word definitions and basic how-to explanations do not count under the description given.
Why do the independent estimates differ?
Ahrefs and seoClarity used different models, time periods and definitions of search-like activity. Ahrefs estimated about 12% in February 2026; seoClarity estimated about 7.5% on Oct. 28, 2026. Neither described its result as a direct measurement of market share.
Does one-fifth of Google’s query volume mean ChatGPT sends one-fifth as much website traffic?
No. Query volume and visits to websites are different measures. In Ahrefs’ October 2026 panel of 118,671 websites, Google accounted for 24.74% of visits and ChatGPT for 0.36%; that panel measures referrals in its sample, not total searches.
What information is missing to verify the claim?
OpenAI did not specify the Google query total, period, countries or underlying data behind the ratio. A verifiable comparison would also need consistent definitions of which ChatGPT questions count as searches and comparable data from both services.
Source: ThorstenMeyerAI.com
