Only 11% of Domains Get Cited by Both ChatGPT and Perplexity — Here Is the Measurement Fix

If you are measuring "AI visibility" as a single number, you are measuring fiction. In Profound's analysis of 100,000 prompts run on both engines, only 11.0% of cited domains appeared in both ChatGPT and Perplexity. Each engine pulls from different source pools, weights different authority signals, and rewards different content structures. Operators treating AI as one channel are blind to which platforms they are winning and losing on — and that blindness is now a budget problem.

Each AI Engine Cites Different Sources — and the Overlap Is Tiny

Profound's data quantified what I have been suspecting for months: the AI citation landscape is not a channel. It is a fragmented archipelago.

In Profound's dataset of 680 million citations (August 2024 to June 2025), Wikipedia accounted for 47.9% of the citations among ChatGPT's ten most-cited sources, while Reddit accounted for 46.7% of those among Perplexity's.

The ChatGPT and Perplexity overlap sits at 11% of cited domains. In Ahrefs' analysis of 730,000 response pairs, Google AI Overviews and AI Mode shared only 13.7% of their citations despite 86% semantic similarity between their answers.

Engines also differ in which sources they cite, so measuring one platform misses much of your citation surface.

Enterprise Measurement Just Arrived — and It Confirms the Problem

Adobe launched Brand Visibility on June 17, combining Semrush's AI visibility intelligence with Adobe Experience Manager. The product draws on nearly 300 million real-world AI search prompts — the largest global dataset of its kind — to show which prompts a brand wins or loses across ChatGPT, Google AI Mode, Microsoft Copilot, and Perplexity.

That Adobe built this tells you everything about where enterprise marketing is headed. As their SVP Anil Chakravarthy put it: "Visibility is everything now" — customers interact with AI tools before ever reaching a website.

The scale backs the urgency. ChatGPT now has 900 million weekly active users (TechCrunch, February 2026). AI traffic to U.S. retail sites surged 1,324% between October 2024 and May 2026.

As Search Engine Land reported, AI visibility starts before search and ends with citations — not clicks. If your measurement stack cannot distinguish ChatGPT visibility from Perplexity visibility from Google AI Mode visibility, you are allocating budget against a composite number that does not exist in the real world.

The Operator Fix: Platform-Specific Citation Tracking

Here is what I am recommending to operators who want to close this gap now, before the enterprise tools mature:

1. Build a per-platform prompt library. Run a fixed set of category-relevant prompts across ChatGPT, Perplexity, Claude, and Google AI Mode on a weekly cadence. Log whether each response cites your domain. Calculate citation frequency as a percentage of total prompts per platform.

2. Track share-of-model, not share-of-voice. Share-of-model measures how often a specific AI engine mentions your brand when discussing your category. This is the metric that replaces share-of-voice in AI search. Segment it by platform.

3. Prioritize brand mentions over backlinks. Ahrefs' study of 75,000 brands found that brand web mentions correlate more strongly with brand visibility in AI Overviews (r = 0.664) than backlinks (0.218).

4. Invest in earned media. Muck Rack's May 2026 "What Is AI Reading?" study found that earned media, sources brands neither own nor pay for, accounts for 84% of cited links in Muck Rack's sample across ChatGPT, Claude, and Gemini. Journalism alone makes up 27%. Paid and advertorial content: 0.3%. The citation economy runs on editorial coverage, not ad spend.

5. Optimize content structure for each engine's preferences. AirOps found that pages with sequential heading hierarchies had 2.8x higher citation likelihood. Treat structure as a pattern to test on your own pages, not a guarantee.

Why This Is a Machine Relations Problem

This fragmentation is exactly the problem Machine Relations was built to solve. Traditional SEO assumed one search engine with one algorithm. GEO extended that to generative engines — but treated them as a single category.

The data now shows that managing brand visibility across AI retrieval surfaces requires platform-by-platform measurement, platform-specific content strategies, and a measurement layer that sits between your content operation and each engine's citation behavior. That is Machine Relations — not as a concept, but as an operational discipline.

The brands that build per-platform citation tracking now will see a measurement advantage for the next 18 months while competitors are still reporting aggregate "AI traffic" as if it means something.

FAQ

Why do ChatGPT and Perplexity cite different sources?

Each engine uses a different retrieval architecture. ChatGPT relies more heavily on its training data and web browsing through Bing's index, favoring Wikipedia and established knowledge sources. Perplexity runs real-time web searches with its own crawling infrastructure, which surfaces more Reddit and forum content. In Profound's sample, only 11.0% of cited domains overlapped between the two platforms.

What should replace CTR as the primary AI visibility metric?

Citation frequency per platform and share-of-model per category. CTR assumes a click happened — but in Semrush's US clickstream study, 92 to 94% of Google AI Mode sessions ended without a visit to any external website. The metric that matters is whether AI engines cite your brand when users ask about your category. Track it per engine, per week, against your top three competitors.

Is earned media really more effective than owned content for AI citations?

Yes, by a wide margin. 84% of cited links come from earned media, the sources brands neither own nor pay for, in Muck Rack's May 2026 study, with paid and advertorial content together at 0.3% in the May 2026 sample. Editorially independent sources are what these engines actually pull from. The operators winning in AI visibility are the ones directing budget toward earned media placement and away from paid promotional content.