AI Citability Checker

Citability is the property of a passage that can be lifted out of your page and dropped into an answer without breaking. An engine generating an answer does not quote pages, it quotes fragments. If a fragment only makes sense in the context of the three paragraphs above it, it cannot travel, and the engine will paraphrase a competitor instead. That is why citability is not the same thing as quality. A brilliant essay written as a continuous argument can be almost impossible to quote, while a plainer page organised into self-contained sections gets cited repeatedly. The tool measures the structural properties that let a fragment stand alone, not whether the writing is good.

Because it is the single defect that most often stops otherwise strong pages from being quoted, and it is nearly invisible to the person who wrote them. A section whose first sentence begins with it, this, however or therefore depends on what precedes it. Extracted on its own, it is unusable. We found this on our own pages before building the tool. A house convention placed the takeaway at the end of each section, which reads beautifully to a human moving top to bottom and leaves an engine with nothing to grab at the point where it looks first. Moving the claim to the opening sentence of each section changes nothing for your readers and everything for extraction, which is why the signal carries twenty-five of the hundred points.

Above eighty is a page an engine can work with comfortably. Between fifty-five and eighty there is usually one structural problem worth fixing, most often sections that open with a dependent sentence or an absence of lists and tables. Below fifty-five the page is likely to be summarised rather than quoted, which means a competitor gets the citation. Do not chase a hundred. The last points often come from adding figures or a question block that the page does not genuinely need, and padding a page to satisfy a checker is exactly the behaviour that makes content worse. Treat the score as a diagnostic that points at a specific weakness, not as a target to optimise.

No. Citability is a necessary condition, not a sufficient one. It makes your page usable by an engine that has already decided to consider you. Whether the engine considers you at all depends on things this tool cannot see: whether your domain carries authority in the subject, whether other sites reference you, and whether your page answers the specific question that was asked. The honest way to read the score is as removing an excuse. A page that scores ninety and is still not cited has a visibility problem rather than a structure problem, and that is useful to know because the two require completely different work.

It counts figures, percentages, dates and amounts in the editorial body, then expresses them per hundred words. A page with a little more than one figure per hundred words carries enough verifiable material to be worth quoting rather than paraphrasing. The reasoning is that an engine has a choice when it uses your page: reproduce your sentence or restate your idea in its own words. It reproduces the sentence when the sentence contains something it cannot safely rephrase, and a specific number is exactly that. Vague claims get absorbed and attributed to nobody. Precise ones get quoted and attributed to you.

It isolates the editorial body first, using the main or article element when the page provides one, and falls back to the body with navigation, header, footer and sidebars removed. Without that step the word count, the paragraph statistics and the heading structure would all be distorted by the template rather than the content. It also strips scripts and styles before counting anything. This matters more than it sounds: a naive analyser counting raw HTML will find headings and links inside serialised application data and report structure that no reader ever sees. Every number the tool reports comes from what is actually rendered as text.