faq-schema

AI answers are questions-and-answers by nature. A page structured as real Q&A gives the model clean pairs to lift — and the FAQPage markup makes the structure explicit.

Steps

1. Fetch the page and understand what it's about and who it's for.
2. Derive the real questions a buyer or reader would ask about this topic — not generic filler. Draw them from: what the page implicitly answers, related searches, and how people actually phrase things (use the buyer-questions skill for deep mining). Aim for 5–8 genuine questions.
3. Write each answer as a self-contained 40–80 word passage, answer-first (see content-capsule). Each answer must stand alone.
4. Generate FAQPage JSON-LD with each question/answer pair.

Output format

First, the human-readable FAQ section (H2 "Frequently asked questions", each question an H3, answer below).

Then the JSON-LD:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "[The question]",
      "acceptedAnswer": { "@type": "Answer", "text": "[The self-contained answer]" }
    }
  ]
}
</script>

Honest note for the user

Google removed FAQ rich results for most sites in May 2026, so don't expect the star-studded search snippet. The value now is different and real: clean Q&A pairs are highly quotable by ChatGPT/Perplexity/Gemini, and the markup helps AI Mode parse them. Ship it for AI visibility, not for Google rich results.

Built by Raphaël Aubry, founder of Howseen (https://howseen.ai). Writing the answers is one-shot. Seeing which of your answers AI actually surfaces — and when a rival's answer replaces yours — is what Howseen tracks. → https://howseen.ai