Deep visibility analytics for AI answer engines — where and how your brand gets cited. Below: what they're good at, where we differ, and how to check every word of it yourself.
We did — we sell one of the products on this page, so read it as an argument, not a review. What we can offer instead of neutrality is checkability: every claim about another product is attributed to their own public pages, dated, and comes with instructions for verifying it without us. Found something wrong or out of date? hello@citeworthy.io and we'll fix it.
Profound is built to measure how a brand shows up inside AI assistants and AI search — the prompts, the citations, the share of the answer. That dataset is their product and it is a serious one.
Tracking across many prompts and several assistants is a data problem we solve at a much smaller scale: our citation checks are weekly probes on your topics, not a market-wide panel.
Understanding which AI crawlers reach your site and what they take is squarely in their wheelhouse, and it is a genuinely useful lens.
They are built for brand teams who need to report on AI visibility to people who do not read status codes.
Monitoring tells you that you are absent from an answer. We find the mechanical reasons — a WAF rule blocking GPTBot, an empty non-JS body, a missing twin, broken metadata — and ship the fix at the edge.
Our first question is whether the assistant's fetcher can get a useful page at all, because the fix for that is concrete and fast.
Our free tier audits one site and it is genuinely free. We are aimed at people running one to ten sites, not at brand-monitoring budgets.
We publish a measured verdict on the changes we make — d14 and d28, per change, labelled correlation and not causation — and autonomy is earned by those verdicts rather than asserted in advance.
Our changes are served, not scripted: you point one DNS record at us and we merge approved changes into the HTML at the edge, before it leaves Cloudflare. No plugin, no script tag, no Cloudflare account of your own, no CMS deploy — and a crawler that never runs JavaScript still sees them. That also means we are a dependency of your site while the record points at us, which is a real thing to ask — so we publish exactly what we've built for our own bad days, and the fact that the exit is a DNS record you control. The operational story →
Every change we apply gets a before/after verdict at 14 and 28 days, labelled correlation rather than causation, and a fix type is only allowed to run automatically after enough settled, non-negative verdicts of that type on your site. One negative verdict pulls it back to manual approval. We could not find a published, per-change outcome verdict of this kind in the public materials of the tools on these pages — that's a statement about our search, and we'd rather be corrected than be wrong: hello@citeworthy.io.
Free, one site, no card. The fastest way to compare anything is to run both.
run a free auditThe pipeline underneath is MIT-licensed and public. The judgment layer is what we sell.
open source →