Ask ChatGPT "which crypto exchange should I use" and it answers with a short list of names. For hundreds of millions of users, that list is the entire research process. Getting onto it is not luck and not an ad buy. We measured it: 1,000 AI answers, ChatGPT included, and the mechanics are consistent enough to act on.
ChatGPT does not rank pages. It composes an answer from sources it already trusts, then names the brands that keep appearing across them. It blends training data with live browsing, so both long-standing presence AND fresh coverage matter — which makes it slower to move than Perplexity. Your homepage is a rounding error; the sources it reads are the game.
Two systems feed a ChatGPT answer, and knowing which one you are fighting changes the play:
In our exchange study, ChatGPT put Kraken first (82 mentions), then Binance (77) and Coinbase (66). The order is not random: it tracks who is most consistently present across the trusted-source layer, and Kraken's learn hub is the second most-cited source in our entire dataset.
Yes, constantly, and more decisively than most teams realize. Across our 20 buyer prompts it named specific brands in the large majority of category answers. The uncomfortable part: it named the same small set over and over. In exchanges, three names filled every top-3 slot and no fourth brand broke in. ChatGPT does not give you a long tail to hide in. You are named or you are absent. See the full study: which crypto exchange AI engines recommend.
Because training data rewards incumbency, the single highest-leverage long-term move is showing up everywhere, consistently, for a sustained period. There is no shortcut around the training-data layer; there is only starting now.
When it retrieves for "best crypto X," it pulls comparison and review sites (in our data: CoinBureau, comparison roundups, CoinMarketCap). Absent from those pages, absent from the answer.
YouTube was the most-cited source in our dataset, 319 citations, more than any website. ChatGPT's browsing surfaces video reviews and explainers. If your category has none about you, you are invisible to that layer.
The only brand domains ChatGPT cited heavily were educational hubs (kraken.com, 246 citations). It cites pages that answer buyer questions, not pages that pitch.
Media citations feed both the training layer over time and the browsing layer now. Pointed PR, on the outlets engines actually cite, is the compounding play.
"Best exchange for beginners" and "safest exchange" pull different names. Find where you are absent, ship against it, re-measure. That loop is the discipline.
If you only have time to optimize for one engine, know the tradeoff. Perplexity is retrieval-first and shows its citations, so fresh, current coverage moves it fastest, and it is the easiest engine for a smaller brand to break into. ChatGPT blends browsing with training data, so it favors brands with durable, long-standing presence and moves slower. Practical sequence: win Perplexity first with current coverage, then let the same consistent presence compound into ChatGPT over the following months. The full Perplexity playbook is in how to rank on Perplexity.
Want to see whether ChatGPT names you, and which sources it cites instead?
Scan my brand free →Longer than Perplexity, honestly. Browsing-driven mentions can shift in weeks as new coverage is indexed; training-data-driven presence changes over model updates and takes months to years. Realistic expectation from client work: first measurable movement on the browsing layer in roughly six to eight weeks, durable presence a two-quarter build. Anyone promising an overnight ChatGPT ranking is selling screenshots.
Get named in the sources ChatGPT trusts and browses: comparison articles, YouTube reviews, credible media, and your own question-answering content. It composes answers from those sources, then names the brands that recur across them.
Yes. In our July 2026 measurement it named a small, consistent set per category, e.g. Kraken, Binance and Coinbase for exchanges, filling every top-3 slot. It rarely names brands outside that consensus.
Because your competitor appears more consistently across the sources ChatGPT trusts, often through a stronger learn hub, more comparison-content presence, or deeper media coverage. The gap is source presence, not usually product.
Yes. SEO earns a ranked link a user might click; ChatGPT ranking earns a mention inside the answer itself. Different levers: earned coverage and consistent mentions beat backlinks, and being quotable beats being long.
Run the same buyer-style prompts on ChatGPT multiple times and track how often you are named and cited. A citeOS audit does this across 5 engines, 20 prompts, 100 observation points, with a free entry scan and published weights.
Growth marketer working in Web3 since its early days. Sagar has helped 75+ crypto and Web3 projects reach their target audiences, and runs Emergence Media. He built citeOS to make AI visibility measurable: 100-observation-point audits across 5 engines with every weight published in the methodology. Check your own citations with the free scan.
Five engines, twenty buyer prompts, 100 observation points. Published weights, reproducible citations, your baseline in about a minute. No card, no signup.
Scan my brand free →