FAQ Schema

FAQ schema is a piece of structured data that marks up the questions and answers on a page, so a machine reading it can tell precisely where a question ends and its answer begins. It is written as a small block of JSON-LD sitting invisibly in the page's code, alongside the visible questions and answers a reader sees.

Its value is the value of a clean seam. A system assembling a reply wants to lift one answer whole, without dragging in the paragraph before it or truncating halfway through. Marked-up questions and answers give it exactly that, which makes such pages unusually easy to quote.

One thing must be said clearly, because it is where this goes wrong. The markup describes content; it does not create it. Questions and answers marked up but not genuinely present on the page are a label over an empty shelf, and that is checkable. Equally, a page that answers real questions plainly in ordinary text is already doing the substantive work — the markup makes it tidier, not truer.

The harder part, and the part worth your afternoon, is choosing the questions. Not the questions you wish customers asked, and not the ones a template supplies. The ones you actually answer on the telephone every week, phrased the way the caller phrases them.

The human translation is a well-organised noticeboard set against a good conversation. The noticeboard helps people find the answer faster. It is worth nothing at all if the answers pinned to it were never true.