Citation corpus
6
verdicts, one per cited page, from "write this" to "you already hold it"
own-it, earn-it, get-listed, competitor-owned, blocked, held — geo/sources.go
Every tool in this category answers the same question: did the engine say your name. Citation Sources answers the question after it. When Perplexity recommends something for a buying question in your category, the answer is assembled out of pages it retrieved, and those pages are listed. We collect them, read them, and say what each one is, whether you are in it, and what it would take to be.
The list is usually uncomfortable reading. Most of it is other people's writing: a Reddit thread from two years ago, somebody's "best tools for X" roundup, a G2 category page. Those pages are doing the recommending, and not one of them shows up in an audit of your own site.
Where it lives: The Citations tab in the dashboard, per product.
Eight types: community thread, listicle, comparison page, review site, documentation, a competitor's own domain, your own domain, or a plain article. The classification runs off the URL's host and path shape, which is deterministic, free, and explainable — three properties that matter more here than catching every edge case.
We fetch the page and look. Present, absent, or unfavourable when the engine's own answer about you came out negative. Pages we could not read stay marked unknown and are never counted as absence, because "we have not looked" and "you are not there" are different facts and only one is worth acting on.
A Reddit thread is not the same job as a G2 listing. Threads get "earn it" (write the page that answers the question properly), third-party lists get "get listed" (pitch the author), directories get "blocked" (reviews move those, writing does not), and reference material gets "own it". The list sorts by verdict, so the pages you can win sit at the top.
GEO score + cited-thread opportunities. On every plan. What changes by plan is how many questions feed the corpus.
| Basic | Growth | Agency |
|---|---|---|
| Included | Included | Included |
Read the full plan comparison.
Try it free
The free version: which sites supply the AI answers in your category, ranked by share of citations, and whether your domain appears in that corpus at all.
AI Citation Source RadarLive demo
See Citation Sources on a real account, with no signup and nothing to configure.
Open the demoFrom the answers themselves. Search-grounded engines return the sources they retrieved alongside the text, and we store that list per question, per engine, per run. We read the structured annotation the API returns rather than scraping links out of the prose, which is what we used to do and got wrong.
Because that is what the engines retrieve for buying questions. In our own July 2026 audit of our category, 10 of the 22 URLs Perplexity cited were Reddit threads. The action is still a page on your own site, written well enough to be cited next to the thread.
Partly. The public citation scan runs three questions on one engine and shows you the first five cited pages and the top five domains, no account needed.
Domain leaderboard
Citation share per host, so you can see the five domains producing half your category's answers instead of scrolling forty URLs.
Citation corpusRank overlap
Per cited page, the Google position for the question it was cited for — and a count of the ones that have no position at all.
FreePublic citation scan
The front door: a real citation corpus for any domain, email-gated, cached for a day, no signup.
Start with the free visibility check to see whether the engines name you today, then run the scan that shows which pages they used instead.