AI engines lift the passage directly under a heading. A page that asks a question in its title and answers it three paragraphs down is quoted less than one that answers straight away. WriteWorks checks the structure of every published page in its crawl: one H1, headings in order, an opening paragraph that answers the title's question, and the schema that page type should carry.
7-day trial · Cancel anytime


This is the working product, not a mockup: the same visibility scores, briefs and published content your team would be looking at. Start a free trial and point it at your own domain, or take the tour of what sits behind the screen.
Late answers, drifting headings and no statement of type.
Articles that warm up for three paragraphs before answering their own headline give AI engines nothing clean to extract from the top of the page.
A missing H1, two H1s from a theme, or an H2 that jumps straight to H4 blurs which passage belongs to which question.
Draft checks help new content. They say nothing about the hundreds of published articles and FAQs written before anyone thought about AI search.
A well-structured FAQ page without FAQPage schema still leaves engines to infer that the headings are questions and the paragraphs are answers.
One H1, headings in order, answer first, the right schema.
Every crawled page except the homepage is checked for a single H1. No H1, or several, is reported with the count, so you can trace it to the template.
On article and FAQ pages, where passage extraction matters most, skipped heading levels such as h2 to h4 are listed. Marketing layouts aren't held to this, because a skipped level there is near-universal and says little.
When an article's or FAQ's H1 is a question, the first paragraph is checked for an answer: it has to exist and share the question's key terms within its first 100 words. If it doesn't, the page is flagged with the problem spelled out.
Product pages are expected to carry Product (or SoftwareApplication), articles Article, FAQ pages FAQPage, category pages ItemList and the homepage Organization with sameAs. Pages missing it are flagged alongside the structure checks.


What each brand fixed, measured before and after with the same audit, in their own words.
“Our job pages used to reach AI crawlers as a "Loading jobs..." placeholder. One rendering fix later, the share AI crawlers can read has gone from 5% to 100%, and every job page now names itself with a proper heading.”
Part of the machine readability pillar.
Question titles answered in the opening paragraph, or flagged.
Exactly one H1 on every page except the homepage.
Skipped levels listed on articles and FAQs.
Product, Article, FAQPage, ItemList and Organization expected by type.
The exact problem: no H1, 2 H1s, h2 to h4, or a late answer.
Open a flagged page in the content editor to fix the opening.
From reading pages one by one to a ranked queue.
Every structure check in the crawl.
Anyone who owns the published pages AI search quotes.
The published articles and FAQs that need a clearer opening.
Content structure for AI search, measured across the site rather than per draft.
Template-level H1 and heading-order problems, with the pages affected.
A queue of pages to rewrite, each with the problem stated.
“We're a booking marketplace, so listings are everything. Since the fixes, pages fully readable without JavaScript went from 84% to 98%, our sitemap dates finally mean something, and our full llms.txt file is real. Over a third of our audited category pages now list their businesses as structured data, with the remaining markets up next.”
Fhresh team
Read the full storyEverything teams typically ask before getting started.
Product, or SoftwareApplication with SaaS mode on.
Article, BlogPosting or NewsArticle, plus a current dateModified.
FAQPage, with each visible question and answer marked up.
ItemList or CollectionPage, so the listed items are stated.
Check your schemaRelated solutions, adjacent use cases, and platform features.
Headings, openings and page-type schema checked on every published page, inside the AI discoverability audit.