What does the extractability check look for?
It assesses how easily a machine could separate your actual content from the surrounding page furniture. It looks at:
- Semantic landmarks — the primary content wrapped in
<main>or<article>, navigation in<nav>, etc. - Content-to-boilerplate ratio — the page isn't mostly menus, sidebars, ads and widgets with a thin sliver of real content.
- Clean separation — that an extraction tool can confidently isolate the article from the chrome.
Main content cleanly isolatable passes; some boilerplate mixed in is a warning; content indistinguishable from nav and ads is a fail.
How is it evaluated, and how is it scored?
GEObubbly runs a readability/extraction pass, checks the content-to-boilerplate ratio, and looks for semantic wrapping of the primary content. It's a core Content Structure & Quality check worth 2 points and runs partially, since judging clean separation depends on how the live page is structured.
Why content extractability matters for GEO
"Extractability" is how cleanly a machine can separate your actual content from the surrounding chrome — menus, sidebars, ads, footers, cookie banners. A reader's eye does this instantly; a parser has to infer it from structure. Two things help most: wrapping the primary content in a semantic landmark (<main> or <article>), and a healthy content-to-boilerplate ratio (the page isn't mostly navigation and widgets with a thin sliver of content). The cleaner the separation, the more confidently an AI engine can lift the right passage. This is increasingly the difference between being cited and being skipped: answer engines quote short, self-contained chunks, and if they can't tell where your content begins and ends, they're less likely to use it. Pairing clean structure with explicit structured data makes the boundary unmistakable, and it builds on a clean heading hierarchy across Content Structure & Quality and GEO / LLM Readiness.