ChatGPT CTR by Position and Country: What the Data Actually Shows (2026)

Short answer: No one has published a single dataset that measures ChatGPT click-through rate (CTR) by ranking position and by country — that study does not exist yet. What we can do is triangulate the real, verifiable numbers that surround the question: how often ChatGPT actually sends a click to a website, what decides which sources it cites (and in what order), and which countries drive its usage. This guide pulls those verified 2026 figures together and labels exactly where each one comes from.

If you are trying to rank — on Google and inside ChatGPT — the honest picture matters far more than a tidy curve someone invented. Here is what the data from Pew Research, Ahrefs, Seer Interactive, Semrush and OpenAI actually shows, and what it means for your AI SEO strategy.

Want to know whether ChatGPT is citing your site at all? Run a free website audit to see where you stand.

Key Takeaways

• There is no verified “ChatGPT CTR by position and country” dataset. Anyone presenting one as a single source is stitching numbers together — so we show our seams.
• AI answers suppress clicks. Pew Research found that when Google shows an AI summary, users click a traditional result just 8% of the time (versus 15% without one), and click a link inside the summary only 1% of the time.
• “Position” in ChatGPT is not Google position. Ahrefs found only about 12% of AI-cited URLs rank in Google’s top 10; Seer found 87% of SearchGPT citations matched Bing’s top results. Getting cited tracks retrieval rank, not your blue-link ranking.
• The US and India dominate ChatGPT traffic (roughly 15–19% and ~9% of visits), but those are browser figures that exclude the apps — so treat country splits as directional.
• ChatGPT is now a real referral channel: Semrush measured referral traffic up 206% year over year, against a backdrop of 800M+ weekly users reported by OpenAI.
• Takeaway for SEO and GEO: optimise to be retrieved and cited, not just ranked — that is a different game from classic SERP CTR.

Does “ChatGPT CTR by Position and Country” Actually Exist?

Let us start with the question everyone types into Google and, increasingly, into ChatGPT itself: is there a table somewhere that says “the first source ChatGPT cites gets X% of clicks in the US and Y% in India”?

No. As of 2026, that combined dataset has not been published by anyone. Here is why it is harder than it sounds. Traditional CTR studies — like Google’s classic position curves — work because a search results page has fixed, numbered positions and a measurable click on each one. ChatGPT’s answers do not work that way. It synthesises a response and may cite a handful of sources, sometimes inline and sometimes in a “references” list, and the user often gets what they need without clicking anything at all. There is no stable “position 1 through 10,” and the click, when it happens, is comparatively rare.

So when you see a confident “ChatGPT CTR by position” chart, be sceptical. The reliable data actually lives in three separate buckets:

1. How often ChatGPT sends any click to a website (referral behaviour).
2. What decides which sources get cited, and roughly in what order (citation studies).
3. Where ChatGPT’s users are (country and traffic data).

Stitch those together honestly and you get a far more useful picture than a made-up curve. That transparency is also, not coincidentally, exactly what earns citations from AI engines: they favour sources that are specific, well-sourced, and honest about uncertainty.

How Often Does ChatGPT Actually Send a Click to a Website?

This is the real heart of “CTR” for AI search. The most rigorous measurement we have of click behaviour in an AI answer comes from Pew Research Center (July 2025), which tracked the actual browsing of 900 US adults across 68,879 Google searches in March 2025. It studied Google’s AI Overviews rather than ChatGPT, but it is the best available proxy for how AI-generated answers change clicking behaviour:

The lesson applies directly to ChatGPT: when an AI gives a complete answer, the click becomes the exception, not the rule. That is the single most important thing to internalise about AI-search CTR — the denominator is tiny.

But “a tiny share of a fast-growing number” is still a real channel. Semrush’s analysis of 17 months of US clickstream data (October 2024 to February 2026) found that ChatGPT referral traffic to websites grew 206% year over year from January 2025 to January 2026. In an earlier Semrush study, the number of unique domains receiving ChatGPT referrals rose 300% in the second half of 2024 alone. The clicks are rarer per answer, but there are rapidly more answers being generated.

Two more findings from that clickstream work shape how those clicks are distributed:

• ChatGPT triggered a live web search on 34.5% of queries (down from roughly 46% in late 2024) — meaning most answers still come from the model’s training data, not fresh citations.
• Over 30% of ChatGPT’s outbound clicks go to just 10 domains, with Google properties alone taking around 21.6%. The referral pie is highly concentrated.

Does Position Matter Inside ChatGPT? What Decides Who Gets Cited

Since there is no clean “CTR by slot” number, the practical version of the question is this: what gets you cited in the first place, and does being cited “first” actually help? Here the data is genuinely useful — and surprising.

The biggest myth is that ranking #1 on Google gets you cited by ChatGPT. It does not. Ahrefs analysed 15,000 long-tail prompts across ChatGPT, Gemini and Copilot (July 2025) and found that only about 12% of AI-cited URLs ranked in Google’s top 10 for the same query — and roughly 80% did not rank in Google’s top 100 at all.

If not Google rank, then what? Seer Interactive (February 2025) looked at 500+ citations from about 100 SearchGPT questions and found the answer: Bing. More than 87% of SearchGPT citations matched Bing’s top organic results, while only 56% matched Google — and the matched Google results had a median rank of 17. It is a small sample, which Seer flags itself, but the direction is clear and consistent with how ChatGPT’s search is built. This is not an accident: Microsoft is both OpenAI’s largest backer and the operator of Bing, and ChatGPT’s web search has historically drawn on Bing’s index. Practically, that means your path into a ChatGPT citation often runs through Bing first.

So the honest translation of “position” for ChatGPT is this: there is no blue-link position — there is retrieval rank. ChatGPT pulls candidate sources (heavily via Bing’s index), then the model decides which to cite based on how directly each one answers the query. “Being first” is less about a numbered slot and more about being the clearest, most quotable answer to the exact question. That is why structured, question-first content gets cited while a page that merely ranks well on Google often does not.

ChatGPT CTR by Country: What We Can and Can’t Say

Country is the second half of the original question, and here the honest answer is that we have usage and traffic data by country, but not click-through-by-country. Still, the usage split tells you exactly where the audience is.

Across providers the pattern is stable, even if the exact percentages wobble month to month:

Two caveats you must carry with those numbers. First, they are browser-visit shares — they largely exclude ChatGPT’s mobile and desktop apps, where a huge and growing share of usage happens, so the true country distribution is fuzzier than any single chart implies. Second, these panels are rolling estimates, so treat them as directional and always note the month they were pulled.

The scale context is what makes the country question matter. At OpenAI’s DevDay in October 2025, CEO Sam Altman said ChatGPT had passed 800 million weekly active users — up from 500 million at the end of March 2025 and roughly 700 million in August. It is a self-reported figure, but even a “rare” click rate, multiplied across 800 million weekly users concentrated in the US and India, is a channel no serious brand can ignore.

The Traditional-Search CTR Baseline (for Contrast)

To understand why AI-search CTR feels so different, compare it with the classic Google curve. The most-cited baseline comes from Backlinko’s analysis of roughly 4 million Google results:

Be careful with that baseline: the underlying data dates to 2022, and other studies disagree — Sistrix puts position 1 nearer 28.5%, while First Page Sage has reported figures closer to 40%. There is no single “correct” position-1 CTR, so always name the source rather than quoting “the” number.

The contrast is the whole point. On a Google results page, position is destiny — more than half of all clicks go to the top three. In ChatGPT there is no top three to win; there is a synthesised answer, a rare click, and a citation list driven by retrieval and answer-quality rather than a numbered rank. Optimising for one does not automatically win the other.

How to Measure Your Own ChatGPT Click-Through (When There Is No Benchmark)

Because no industry benchmark exists, the only CTR that truly matters is your own — and you can measure it. The catch is that Google Search Console will not show it and ChatGPT does not publish per-source analytics, so you have to assemble the picture yourself:

• Isolate AI referral traffic in GA4. Filter sessions whose source or referrer is chatgpt.com (and perplexity.ai, gemini.google.com, and so on). That tells you how many clicks ChatGPT actually sent you — the numerator of your real CTR.
• Watch Bing, not just Google. Because citations track Bing so closely, Bing Webmaster Tools is your early-warning system for AI visibility; rising Bing impressions often precede ChatGPT citations.
• Track citations, not only clicks. Given the roughly 1% click reality, a citation that is never clicked still builds brand awareness and trust. AI-visibility tools — and plain manual prompt testing — let you see whether you are being named even when no click follows.
• Check your server logs for AI crawlers. Hits from GPTBot, OAI-SearchBot and PerplexityBot are the leading indicator that your content is being ingested at all; if they never visit, no citation is possible.

Put together, these give you a defensible “AI-search CTR” for your site — which is the only version of this metric that is both real and actionable today.

What This Means for Your SEO and GEO Strategy

If you take one thing from the verified data, make it this: stop chasing an imaginary “ChatGPT position” and start optimising to be retrieved, quoted and cited. Concretely:

• Win Bing, not just Google. Because ChatGPT’s search leans heavily on Bing’s index (Seer’s 87% match), your Bing Webmaster Tools setup and Bing rankings are no longer optional.
• Answer the exact question, early. Give a direct, quotable answer in the first sentence of a section, then expand. That is what AI engines lift, and it is the core of Answer Engine Optimisation (AEO).
• Build citable, sourced content. Pages that cite primary data (as this one does) and state facts cleanly are far more likely to be cited back — the heart of Generative Engine Optimisation (GEO).
• Structure for machines. Clear headings, FAQ blocks, schema and entity-consistent information help both crawlers and models. If you are invisible today, start with why your website might not be showing up on ChatGPT.
• Think “search everywhere.” Google, Bing, ChatGPT, Gemini and Perplexity each choose sources differently, so visibility now means optimising across every surface, not one SERP.

For the step-by-step playbook, read how to get your brand cited by ChatGPT, Gemini and Perplexity, and if local visibility matters to you, see how to optimise your Google Business Profile using Gemini and ChatGPT.

How SEO Circular Helps You Earn ChatGPT Citations

Ranking in ChatGPT is a different discipline from ranking on Google, and it rewards firms that understand both. At SEO Circular, our AI SEO services are built around getting your brand retrieved and cited across AI answer engines — combining LLM SEO, AEO and GEO with the technical and content foundations that still drive traditional search. We focus on the inputs the data says matter: Bing visibility, quotable question-first content, entity and schema consistency, and durable authority signals.

The honest truth from all of this research is that AI search is still early and the measurement is messy. The brands that win are the ones building citable authority now, while the curves are still being drawn.

Turn AI Search Visibility into Real Traffic

You cannot optimise a “ChatGPT position” that does not exist — but you can absolutely optimise to be the source ChatGPT reaches for. Want to know whether AI engines are citing you today, and where the gaps are?

Start with a free website audit, or talk to our team about an AI-search visibility strategy tailored to your market.

Frequently Asked Questions