Definition
Extractability is how easily a machine can lift a clean, correct, self-contained answer out of a page. Crawlability decides whether a bot can read the page at all; extractability decides whether anything on it is quotable once read. A page can be perfectly crawlable and still give an assistant nothing to use.
At a glance
- CategoriesContent
- Related fieldsContent, GEO
- DifficultyIntermediate
At a glance
- Extractability is judged on the passage an assistant pulls, not on the whole page.
- A self-contained paragraph survives extraction. One opening with “as mentioned above” does not.
- Extractability sits downstream of crawlability and upstream of being cited.
What gets lifted is a passage, not a page
An assistant does not quote your page. It quotes a chunk of it.
The retrieval step splits documents into passages, scores each against the question and pulls back a handful. Content chunking is that split seen from your side, and the boundaries usually land on headings.
So the unit under judgment is a section. If your answer runs across three paragraphs with a table wedged in the middle, the passage that comes back carries a third of it and reads as incomplete.
Two things break a passage. Back-references pointing at content no longer attached. And facts living in a format the extractor cannot read: text inside an image, a specification in a PDF, a table assembled by JavaScript after load.
The page that ranks and never gets quoted
Everyone checks whether they rank. Almost nobody checks whether they are quotable, and the second one is where the quiet losses are.
Take an ordinary pricing page. It ranks well, the copy is fine, and the prices sit in a component that loads after the HTML. A shopper asks an assistant what the product costs. The assistant reads the page, finds no figure it can attribute, and quotes a review site that printed the number in plain text.
You were crawled. You were read. You were skipped, and a third party got named as the authority on your own pricing.
The same thing happens to specifications hidden behind tabs and ingredient lists set as images. Nothing is broken by any conventional measure. There is just nothing a machine can lift with confidence.
Make one template easier to lift from
Pick your highest-value template and work through it once.
- Put a direct answer in the first 60 words under each heading. Full sentence, subject named, no run-up.
- Delete every back-reference. “As noted above” and “this approach” assume context the extractor will not have.
- Print the facts as text. Prices, dimensions, ingredients, compatibility. If it lives in an image or a PDF today, put it on the page as well.
- Give each section one subject. A heading covering two topics produces a passage that answers neither cleanly.
- Read one section aloud with the rest of the page covered. If it still makes sense, it is extractable.
This is editing work rather than engineering work, and one person can do a template in a morning. Answer-first content is the habit that stops it drifting back.
The same facts, extractable or buried
| The fact | Extractable version | Buried version |
|---|---|---|
| Price | Printed in the HTML as text | Loaded by a script after render |
| Specification | Listed under a clear heading | Inside a downloadable PDF |
| The answer | First sentence of the section | Third paragraph, after the context |
| A claim | Stated once with a named source | Implied across two sections |
Frequently asked questions
-
Is extractability the same as crawlability?
No. Crawlability asks whether a bot can fetch your page. Extractability asks whether what it fetched holds a passage worth quoting. They fail for different reasons and get fixed by different people: crawlability is engineering, extractability is editorial.
-
How do I check whether a page is extractable?
Cover everything except one section and read it. If it stands alone, answers one question, and names its own subject rather than saying “it” or “this”, an extractor can use it. Then view the source and confirm the facts you care about are in the HTML rather than added later by a script.
-
Will making a page extractable get it cited?
Not by itself. Extractability removes the reason an assistant would skip you. Being chosen still depends on whether the passage says something specific, current and worth repeating. An extractable page with nothing to say is a clean, quotable nothing.
-
Do PDFs hurt extractability?
They limit it. A PDF can be read, but its contents are harder to chunk cleanly and the page around it carries none of the meaning. If a spec sheet holds facts you want quoted, publish those facts as HTML and keep the PDF as the download.