FAQ pages are the most extraction-friendly format on the web: each entry is a question, a direct answer, and nothing else, which is precisely the passage shape answer engines quote. Most FAQ pages still fail at it, because they are written as marketing rather than as answers. Here is the format that survives extraction.
Write the real question as the heading
Use the wording people actually ask, in full: "How long does shipping take?" rather than "Shipping". Engines match retrieval against question-shaped queries, and a heading that mirrors the query is the strongest relevance signal a passage can carry. Pull the phrasings from your search console queries, support tickets, and the "people also ask" boxes on your topic.
Answer in the first sentence
The sentence directly under the heading must contain the complete answer: a number, a date, a yes or no with the condition. Context, exceptions, and links come after. An extractor takes the first quotable sentences; if those sentences are throat-clearing, the citation goes to a competitor whose first sentence is the answer.
Formatting that survives extraction
- One question per heading; never bundle two questions into one entry.
- Keep each answer self-contained; "see above" breaks when the passage is lifted alone.
- Use real text, not accordions rendered empty until clicked; content hidden from the initial HTML is invisible to most crawlers.
- Add FAQPage structured data that mirrors the visible entries exactly, as covered in our schema guide.
Where FAQ sections belong
Besides a standalone FAQ page, a short FAQ section at the end of an article catches the long-tail questions the article raises. Keep it honest: three to five questions the piece genuinely left open, not a keyword dump. Engines quote FAQ sections from article pages at least as often as from dedicated FAQ pages.
This format is AEO in miniature: question, direct answer, clean structure. The full version of that argument is in What is answer engine optimization?
