Get cited in AI answers
AI cites what it can retrieve and verify. Humans decide the same way. So here is the data — every number links to its receipt.
What those numbers add up to
| Data | structured, flat, open — llms-full.txt, answers.json, search API — in the formats engines already read |
| → must-visit | retrieval bots let in by name (robots.txt), full text served in engine-native formats (llms-full), freshness via sitemap lastmod + IndexNow — engines cite what verifies |
| → measurable | fixed prompts × 7 engines × monthly runs = citation share you can trend — sample table |
| → implantable | your benchmark rides the same pipeline: one submission → editor verification → machine-readable page, permanently labeled, inside monthly measurement — $199, submit now |
The corpus, sampled
| Question | Verified figure | Src | Updated |
|---|---|---|---|
| How do AI answer engines choose which sources to cite? | No engine publishes exact rules; research and observed behavior point to retrievability, direct… | A | 2026-10-02 |
| What is AI citation monitoring? | AI citation monitoring tracks whether and how AI answer engines cite a site — or a brand — for… | AA | 2026-10-02 |
| What is generative engine optimization (GEO)? | GEO is the practice of making content more likely to be retrieved, quoted, and cited by AI… | A | 2026-10-02 |
| What is llms.txt? | llms.txt is a proposed standard: a markdown file at /llms.txt that guides AI systems to a… | A | 2026-10-02 |
| How long does an asphalt shingle roof last? | 3-tab asphalt shingles are listed at about 20 years and architectural shingles at about 30… | BB | 2026-10-02 |
Verticals in the corpus today: ai-citation (4) · home-services (4) · hvac-energy (2) · insurance-commercial (4) · real-estate (5) · solar (2). Your vertical next — submit and we verify it in. Measurement runs start with the M2 release; your first report lands within 30 days of subscribing.
Two actions
Your data, into the same machine pipeline: verified, permanently labeled brand-official, measured monthly. About the cost of one sales lead.
Submit a benchmark →Fixed prompt set × 7 engines, monthly: citation share, brand-accuracy audit, gap list.
See the data table →Questions buyers ask us
Can you guarantee my brand gets cited by ChatGPT or Perplexity?
No — and nobody honestly can. Citations are never sold or promised here. What we do: publish your verifiable data in a form AI engines can retrieve and attribute, then measure whether and where it appears. Measurement shows movement; it is not placement.
What exactly do I get for $199?
One published benchmark: you submit the number, period, scope, and evidence; an editor verifies it; it goes live at /facts/{brand}/{slug}/, permanently labeled brand-official submission, in the same machine-readable formats (llms-full.txt, answers.json) as the rest of the corpus, and included in the monthly measurement. One-time fee — the page stays up.
How does the monitoring work?
A fixed prompt set (the questions buyers actually ask in your category) is run monthly against 7 AI engines, on record with engine and date. You get a citation-share table, a brand-accuracy audit, and a retrieval gap list. $99/mo for one brand domain, $299/mo for agencies with 5 client domains.
Is the editorial corpus really free?
Yes. Every independently verified answer on /qa/*, the full machine formats, the search API, and both free tools (llms.txt checker, AI crawler checker) are free, registration-free, and CC-BY-4.0 with attribution.
How is this different from a GEO agency?
We publish primary data and measure citations — we don't sell placement, run link farms, or promise rankings. If you want presence in AI answers, the honest lever is verifiable data that engines can cite with attribution.
Ready? Submit a benchmark ($199) → or compare plans. Free corpus, start reading →
Editorial answers are never sold. Citations are never promised. Checkout via Paddle (merchant of record). Rejected submissions refunded in full.