Stage 4: Retrieve

Citability check

A per-section checklist of answer-first passages on one page, with the exact passage quoted rather than a score, to show what a system could lift.

Free to use. No login, nothing stored.

What it checks

Retrieve
  • The opening sentence of each H2 and H3 section answers the heading
  • A definition is present when the heading asks "what is"
  • Numbers carry units and a nearby source link
  • Author, published and updated dates are visible
  • FAQ answers are real text, not only markup

Method question for this stage: Can a passage be lifted as a clear answer?

Run the check

We fetch the page once (10 second limit, first 1 MB) and check up to 25 H2 and H3 sections from the server HTML. Nothing is stored.

Results

No check run yet

  • Info

    One checklist per H2 or H3 section, then page-level rows

    Answer-first opening sentence (quoted), definition when the heading is a question, numbers with a nearby source link; then dates, author and FAQ text for the page. Counts only, no score.

How we check

The rules this tool tests, and where they come from

A result is a check against a published rule, not a prediction. Every row shows its working.

Rules and sources

Links go to the operator's or standards body's own documentation. We do not link to third-party summaries.

Read more

Next step

Fix what it finds

This check belongs to the Retrieve stage of the Crawl-to-Citation Method. The service below covers the fix; the Diagnostic ranks it against everything else.

Most engagements start with the Search Visibility Diagnostic ($2,500).

It covers all six method stages, so what this tool finds is ranked against everything else before you commit to any work.

Book the diagnostic

All free tools

Citability questions

What the checklist tests, and what it does not claim.

Still have questions? Book a call

Why is there no score?

A score would imply we can predict whether a passage gets quoted, and nobody outside the engines can. What we can do is test the passage against the Retrieve-stage checklist we use in client work: does the section open with the answer, does a question heading get a definition, do numbers carry a source, is the page dated and signed. Each row quotes the passage so you can disagree with the heuristic.

What makes a passage answer-first?

The first sentence under a heading repeats the heading's key terms and stays short, 40 words or fewer here. A retriever that splits the page into passages then has a self-contained chunk that matches the question. The threshold is ours, chosen so that a two-line answer passes and a paragraph of preamble does not; the quoted sentence lets you judge borderline cases.

Why check dates, authors and FAQ text at page level?

Google's helpful-content guidance asks whether a page shows who wrote it and when, and its AI features documentation says content needs to be in the HTML to be used. An FAQ whose answers exist only in JSON-LD or arrive by JavaScript is not visible text. These rows report what is in the server HTML; they do not judge the quality of the author or the date.

Last reviewed