Kuroma
Question Coverage & FAQ Presence
What this check reads on your page
Three kinds of question evidence are read here. Any FAQPage JSON-LD is parsed and credited only when its question list is non-empty, and the markup is searched for an FAQ section. The headings from h1 to h4 are read, counting those that contain a question mark, in either the ASCII or the full-width form, or that open with an English question word. Sentences in the visible text that end in a question mark are counted as well, and English question sentences such as 'what is' or 'how do' are pulled out to show you which questions the page already poses.
Our verdict on the evidence
Evidence-backed Covering the sub-questions users actually ask is the single strongest diagnosed lever: contextual gaps and intent divergence explain most zero-citation failures. FAQ formatting by itself is not the lever; answering the questions is.
Zero-citation failure taxonomy: contextual gap, intent divergence, information scarcity (arXiv 2603.09296) · Google Search Central guidance on AI features (no special schema needed, no ideal page length)
AI engines match a page to the question being asked, and a page that never addresses the question at all is among the most common diagnosed causes of going uncited. Marking the questions you answer is how a retriever finds the passage for each of them.
What it is worth
26% of On-Page Content, which is 6.5 points of the overall 100.
Not scored on data pages (rankings, statistics and paginated indexes): the factor is dropped from the weighted sum and from the denominator, so a page of that type is never charged for it.
On a page written in Japanese, Korean or Traditional Chinese, only the parts of this check that work in any language are scored, over a budget of 85 rather than 100, and the result is rescaled to 100 so it stays comparable.
How audited pages score on it
Based on 8,352 audited pages.
The median page scores 9, and the middle half of pages score between 0 and 34.
- Good (75 to 100)457 (5%)
- Moderate (50 to 74)796 (10%)
- Needs work (25 to 49)2,056 (25%)
- Critical (0 to 24)5,043 (60%)
85% of audited pages score below 50 on it.
Where it fails most and least
Industries that fail it most
- Marketplace: 100% of 43 audited pages fail it
- Automotive / Mobility: 97% of 151 audited pages fail it
- Gaming: 96% of 192 audited pages fail it
- Food & Beverage / CPG: 96% of 180 audited pages fail it
- Pharma / Biotech: 96% of 132 audited pages fail it
Industries that fail it least
- AI / Machine Learning: 62% of 398 audited pages fail it
- Mental Health / Wellness: 63% of 52 audited pages fail it
- HR Tech / Recruitment: 72% of 69 audited pages fail it
- MarTech / Marketing Platform: 73% of 108 audited pages fail it
- SaaS / Cloud Software: 74% of 131 audited pages fail it
How to fix it
- Write the heading of each main section as the question a reader would type, and end it with a question mark; the heading tier is full at five.
- Give the FAQ block a class or id that contains faq, so the section is recognised even without JSON-LD.
- If the page has a FAQ, mark it up with FAQPage JSON-LD that carries the real questions and answers in mainEntity; an empty list scores nothing.
- Cover the follow-up questions a buyer asks after the first one, so a dozen question sentences appear in the text rather than one.