FAQPage schema is the Schema.org structured data type that marks a web page as a list of frequently asked questions with their answers, so that search engines can identify each question and its accepted answer without interpreting the layout. In structured data for SEO, it is usually written in JSON-LD and is also known simply as FAQ schema.
For several years, FAQPage markup could produce expandable question rich results directly below a search listing. Since August 2023, Google shows these rich results only for a limited set of well-known, authoritative government and health websites, so for most sites the markup now serves machine understanding rather than visible search enhancements.
How FAQPage schema is structured
The markup follows a simple nesting. A FAQPage object contains a mainEntity property listing one or more Question objects; each Question has a name holding the question text and an acceptedAnswer holding an Answer object whose text contains the answer. The table summarizes the required elements.
| Property | Belongs to | Content |
|---|---|---|
| @type FAQPage | Page-level object | Declares the page as an FAQ |
| mainEntity | FAQPage | Array of Question objects |
| name | Question | The full question text |
| acceptedAnswer | Question | One Answer object |
| text | Answer | The full answer, limited HTML allowed |
Answer text may include a small set of HTML elements, such as links, lists, paragraphs and emphasis, which Google can display when the rich result is eligible. Other tags are ignored. The answer in the markup must match the visible answer on the page; structured data describes content, it does not replace it.
A minimal JSON-LD block declares the Schema.org context and "@type": "FAQPage", followed by the mainEntity array. It can sit anywhere in the HTML, including content injected by JavaScript, as long as the rendered page contains it. Only one FAQPage object should describe a given page; several separate blocks for the same FAQ create conflicting or duplicate entities.
Eligibility and content rules
Search engines apply content policies on top of syntax. Valid markup on an unsuitable page is ignored or can trigger a structured data manual action. The essential conditions are clear and stable.
- The site provides both questions and answers. Pages where users submit answers belong to QAPage instead.
- Every marked question is visible on the page, either directly or in an expandable section.
- No advertising purpose. Questions written as promotional copy are not appropriate FAQ content.
- No offensive, dangerous or misleading content in questions or answers.
- One instance per question. Repeating the same FAQ across many pages adds no value and can look manipulative.
Since the 2023 change, being eligible under these rules no longer guarantees any visible feature for ordinary commercial or editorial sites. Google has stated that it does not penalize sites for keeping the markup, so existing FAQPage blocks can stay in place.
FAQPage compared with related schema types
Several Schema.org types describe question-and-answer content, and choosing the wrong one is a frequent validation and policy error. The decisive question is who writes the answers and how many answers each question has.
| Schema type | Use when | Example page |
|---|---|---|
| FAQPage | The site writes several questions and one answer each | Shipping and returns FAQ |
| QAPage | One question with answers from users | Community forum thread |
| HowTo | Step-by-step instructions | Assembly guide |
| Article | Editorial content with a headline | News report with an FAQ section |
| Quiz | Educational questions testing knowledge | Flashcard practice page |
A page can combine types. An article with a short FAQ at the end can carry Article markup for the page and FAQPage markup for the question section, provided both describe visible content. Google also reduced the visibility of HowTo rich results in 2023, so neither type should be chosen for display effects alone.
Why FAQPage schema still matters
Even without rich results, FAQ markup gives machines an explicit question-to-answer mapping. This helps search engines and answer engines extract precise responses, and it encourages editors to write self-contained answers that can be quoted on their own.
| Site type | Rich result likelihood | Main remaining benefit |
|---|---|---|
| Government and health authorities | Possible | Visible expanded answers in search |
| Ecommerce and service sites | Very unlikely | Clear machine-readable support content |
| Publishers and blogs | Very unlikely | Structured answers for extraction |
| Help centers and documentation | Very unlikely | Consistent structure across many articles |
The effect on AI answer engines is not formally documented. Generative systems mainly read the visible text, so the clearer answer wording usually matters more than the markup itself. Treat FAQPage as a low-cost supporting signal, not as a ranking lever.
Well-structured FAQ content also supports internal search and support deflection: the same questions can feed a help widget, a chatbot knowledge source and the page markup from one maintained list.
Implementation steps and common mistakes
A reliable implementation generates the markup from the same data that renders the visible FAQ, so the two cannot drift apart. Validation should then use both the Rich Results Test for Google-specific eligibility and the Schema Markup Validator for general Schema.org correctness.
| Mistake | Consequence | Correction |
|---|---|---|
| Hidden or missing questions | Policy violation, markup ignored | Show every marked question |
| Answers differing from page text | Misleading structured data | Generate both from one source |
| FAQPage on forum threads | Wrong type for user answers | Use QAPage |
| Same FAQ on every page | Duplicate, low-value markup | Mark it up on one page only |
| Expecting traffic gains | Disappointment after deployment | Measure as content quality work |
Content management systems and SEO plugins often add FAQ markup automatically through FAQ blocks. This is convenient, but it can mark up content that is not a real FAQ, such as sales objections or marketing claims phrased as questions. A periodic audit of all pages carrying FAQPage, with a check that each question answers a genuine user need, keeps the markup accurate and within policy.
A plain-language summary
FAQPage schema is a hidden label added to a page of questions and answers. People reading the page do not see it, but search engines do. It tells them, in a fixed format, this is a question and this is its answer.
In the past, the label often made answers appear directly in Google results. Today that happens mainly for official government and health sites, yet the label still helps machines understand the page precisely.
An example from practice
A regional public health agency publishes a vaccination page with twelve questions, from eligibility to side effects. Its web team generates FAQPage JSON-LD from the same content database that renders the visible accordion, so every answer in the markup matches the page exactly.
The Rich Results Test confirms eligibility, and because the agency is an authoritative health source, some queries show expandable questions below its listing. A furniture retailer using the same technique on its delivery FAQ gets no rich result, but the clean structure still gives search engines and assistants precise, quotable answers about delivery times and return windows.