Free tool · Flagship
Citability scorer
Paste a page and get a 0–100 citability score against the signals AI answer engines actually reward — answer-first structure, question headings, concrete stats, tables, lists, FAQs, freshness and sources — with specific fixes for what's weak. It all runs in your browser; your content is never sent anywhere.
—
Paste some content to see your score and the prioritized fixes.
Prioritized fixes
Requested → shippedİpek asked for a way to paste two pages and see exactly which signals flipped — via Product Hunt. Here it is.
Paste content into both panes to see which page scores higher and exactly which signals flipped.
Signal-by-signal diff
All 11 checks, points each side earned, sorted by the biggest gap first.
What it scores, and why
These aren't invented rules — each maps to how AI answer engines assemble and cite content, and to the argument in my own get-cited playbook and where to focus.
- Answer-first opening — engines lift the direct answer near the top. A tight opening paragraph that answers the question wins over a slow windup.
- Question-style headings —
H2s phrased as real questions match how people prompt, so they're easy to quote verbatim. - Concrete stats & numbers — specific figures are the most quotable, citable unit of content.
- Lists, tables & an FAQ — structured chunks are easy for a model to extract cleanly.
- Scannable paragraphs & real depth — short paragraphs and enough substance to actually answer.
- External sources & a freshness signal — citations to authorities and a visible date signal trustworthiness.
It's a heuristic guide, not a guarantee — no tool can promise a citation. It points you at the structural gaps that keep good content from being quoted.
Curious how the pros score? See the AI citability benchmark — 17 well-known SEO, marketing and AI pages run through this exact scorer, with what the top scorers do differently.
How the score is calculated (the exact weights)
Eleven checks sum to 100 points. Each returns partial credit, so a decent-but-imperfect page still scores fairly. Nothing is sent anywhere — the analysis runs entirely in your browser, on the text you paste, using the weights below:
- Answer-first opening16
- Question-style headings14
- Concrete stats & numbers12
- Clear heading structure8
- Uses lists8
- Has an FAQ section8
- Scannable paragraphs8
- Enough depth (word count)8
- Includes a table6
- Cites external sources6
- Freshness signal6
Answer-first structure and question-style headings carry the most weight because they map most directly to how retrieval-augmented answer engines chunk a page and match it to a prompt. The checklist re-sorts by how many points each fix would recover, so the top item is always your biggest win.
Frequently asked questions
What makes a page "citable" by AI?
AI answer engines favour pages that answer the question directly and early, use question-style headings that match how people prompt, back claims with concrete numbers and sources, and organise information into extractable chunks (lists, tables, FAQs) with short, scannable paragraphs and a visible freshness signal. This tool scores a page against exactly those signals.
Is the citability score guaranteed to get me cited?
No — and any tool that promises that is overselling. Citations depend on factors no on-page tool can see: your domain's authority, third-party mentions, and each engine's changing retrieval. The score measures the structural signals you control, which are necessary but not sufficient. Treat a high score as removing the on-page reasons not to cite you.
Does my content get uploaded anywhere?
No. The entire analysis runs locally in your browser with JavaScript — your pasted text never leaves your device and nothing is stored.
Should I paste Markdown or plain text?
Either works, but Markdown (or the real HTML) gives the most accurate score because the tool can see your headings, lists, tables and links. If you paste plain text with no structure, the structural checks will read low — which is often the honest result.
How is the score calculated?
Eleven weighted heuristics sum to 100: answer-first opening (16), question-style headings (14), concrete stats (12), heading structure, lists, an FAQ section, scannable paragraph length, real depth by word count (8 each), a table, external authoritative links and a freshness signal (6 each). Each check contributes partial credit and a specific fix when it is weak, and the checklist re-sorts so your highest-impact fix is always on top.