1 — The answer is a shortlist of ~4 brands. There is no page two.
When a buyer asks "what's the best X", AI doesn't return links — it names about four brands and moves on (measured across 421 answers in our shortlist study). If you're not one of them, there's no ranking to climb and no click to lose: the deal is over before your analytics see anything.
2 — AI reads sources, not your website.
In every category we've measured, the pages feeding AI answers are comparison listicles, Reddit threads, YouTube videos and review sites — not official websites. In payments, two rival companies' blogs were cited on more questions than stripe.com itself. In AI coding tools, cursor.com appeared on 1 of 18 questions while a competitor's blog covered 10. Your homepage isn't in the conversation; the pages about your category are.
3 — Your visibility is different on every engine.
The same brand, same questions: 93% recommendation share on one engine, 53% on another. A well-known brand at 5% on ChatGPT and 69% on Google AI Overviews. Each engine reads different sources and shows measurable leanings — so "does AI recommend us" has no single answer. It has one answer per engine.
4 — And it moves: ~15% of answers flip on re-ask.
Across three published studies, about 15% of identical question × engine checks changed verdicts between runs. A single screenshot of "what ChatGPT says about us" is close to meaningless; visibility is a number you track, not a fact you check once.
What actually changes the number
1 · Find the doors. Pull the exact sources AI cites for your category. The ones that don't name you — especially "best X" comparison lists — are warm doors: AI already trusts them, they just don't mention you yet. Pitching those beats publishing new content.
2 · Publish where AI reads. In every category we measured, community threads and video out-covered everyone's official blogs. If your category's answers are fed by YouTube and Reddit, another article on your own domain won't compete.
3 · Re-measure weekly. Answers drift, competitors move, sources update. The teams that win treat AI visibility like a metric — baseline, action, re-measure — not a one-time audit.
Where these numbers come from
All figures are from Ainswer's published multi-run studies — SaaS payments (235 answers), the GEO industry (186), and AI coding assistants (210) — each check repeated and majority-scored across ChatGPT, Perplexity, Google AI Overviews and Claude. Methodology details are in each report.
Stop guessing. See your number.
Paste your site — one real buyer question across five engines, free, no email. The full scan shows your share, your engines, and the exact sources to fix.