FAQ schema is a small piece of code that labels your question-and-answer pairs so machines know exactly what they are. It does not change what a human sees. It tells AI search "these are FAQs," which makes them easy to lift and credit. Pages with FAQPage schema are quoted more often in AI answers. It is one of the highest-return, lowest-effort moves you can make.
I run a social media and education business in New Zealand. FAQ schema is the closest thing to a free win I have found in AI search. You write the questions and answers anyway. The schema just labels them for machines. Our SEO partner, The Optimisers, treats it as a default on every page. This guide explains what it is, the citation lift it gives, and exactly how to add it.
What FAQPage schema actually is
FAQPage schema is structured data, written in a format called JSON-LD, that sits in your page's code and labels a set of question-and-answer pairs. Humans never see it. Machines read it and understand "this block is an FAQ, here is the question, here is the answer." That clarity is what makes the content easy to lift.
Think of it as a label on a box. The box is your FAQ content. Without a label, a machine has to open it and guess what is inside. With the label, it knows instantly. It does not have to interpret your layout or guess which line is the question and which is the answer.
This is part of a wider family of schema types: Article, FAQPage, HowTo, Organization, and Person. FAQPage is the one tied most directly to AI citations, because its shape matches how people ask AI questions.
Why it lifts your citations
FAQ schema lifts citations because it hands the machine a perfect match. People ask AI full questions. Your FAQ is already a stack of questions with direct answers. The schema confirms that shape so the AI can lift a pair whole and credit you. The cleaner the match, the higher your odds of being the quoted source.
Pages with FAQPage schema show measurably higher citation rates in AI answers, which is why our SEO partner adds it to every page as standard. The reason is simple. A labelled question-and-answer pair is the single easiest thing for a machine to extract with confidence.
Remember the zero-click reality. About 65 percent of searches now end with no click, according to widely reported 2026 data. The AI answer is the whole game. FAQ schema makes you more likely to be the source inside that answer, which is the only win left when nobody clicks.
What it does not do
Schema is not magic and it is not a ranking shortcut. FAQ schema does not improve weak content. It labels good content so machines read it cleanly. If your answers are vague or your page has nothing worth quoting, the schema labels emptiness. The content has to be strong first.
It also does not change what your reader sees on the page. The schema lives in the code. Your visible FAQ section is separate. You need both: the visible questions and answers for humans, and the schema labelling them for machines.
So treat it as an amplifier, not a fix. Write real questions, answer them directly, then add the schema to make sure the machine never has to guess.
How to write the FAQ first
Write the FAQ for humans before you touch any code. The schema only labels what is already there. Start with the real questions people ask you, answer each one directly in 40 to 60 words, and put the section at the foot of the page. Then the schema has something worth labelling.
Here is the writing pass.
- List the real questions. Pull them from what customers ask, from "people also ask" boxes, and from how people phrase prompts.
- Use spoken phrasing. Full questions, the way people actually say them out loud.
- Answer directly. First sentence is the answer. No wind-up.
- Keep each answer to 40 to 60 words. Long enough to be complete, short enough to lift whole.
- Include four to six pairs. Cover the genuine follow-up questions, not filler.
How to add the schema
Adding the schema means placing a small JSON-LD block in your page's code that lists each question and its answer. Most modern CMS platforms have a field or a plugin for this, so you often do not touch raw code at all. If you do, it is one block in the page head, and it is the same shape every time.
Here is the shape of the code, in plain terms.
- A wrapper that says "this is an FAQPage."
- For each pair: a "Question" with the question text, and an "Answer" with the answer text.
- The text in the schema should match the visible text on the page exactly.
How to ship it.
- Check your CMS first. Many have an FAQ block that adds the schema for you.
- If not, ask your developer to drop the JSON-LD block in the page.
- Make the schema text match the on-page text word for word.
- Test it with Google's Rich Results Test to confirm it reads.
- Add it to every page with an FAQ, then make FAQs standard on new pages.
A simple rollout for a Kiwi business
For a small New Zealand business, the whole job is a single focused week. You probably already have FAQs scattered across your site, or questions your team answers by email every day. Gather them, shape them into clean pairs, add the schema, and you have lifted your citation odds across your whole site without writing a single new article.
Run it like this.
- Pick your ten most important pages. Service pages, key blog posts, your homepage.
- Add a four-to-six-question FAQ to each, written from real customer questions.
- Add FAQPage schema to all ten.
- Make FAQ-plus-schema the default for every new page from now on.
- Track which questions start showing your brand in AI answers, then add more like them.
This is reachable without a big budget. The questions are ones you already answer. The schema is a one-time code addition that keeps paying off. It is the kind of move that quietly compounds.
Frequently asked questions
Do I need a developer to add FAQ schema?
Often no. Many modern CMS platforms, including most built on Next.js or with SEO plugins, have an FAQ block or a schema field that adds the JSON-LD for you. If yours does not, a developer can add it in minutes. It is one small block of code per page, the same shape every time.
Will FAQ schema get me a rich result in Google too?
It can, though Google has narrowed FAQ rich results to certain site types. Even when the visible rich result does not show, the schema still helps machines read your content cleanly, which feeds AI citations. So add it for the AI-search benefit regardless of whether the Google rich snippet appears for you.
Does the schema text have to match my visible FAQ exactly?
Yes. The text in your FAQ schema must match the question-and-answer text shown on the page. Mismatched or hidden schema content breaks the rules and can get you penalised. Write the FAQ for humans, then mirror that exact text in the schema. Same words in both places, every time.
How many FAQs can I mark up on one page?
There is no hard limit, but quality beats quantity. Four to six strong pairs that answer real questions will outperform twenty filler ones. Each pair should come from a genuine prompt and answer it directly. Padding the list with weak questions just dilutes the strong blocks the machine wants to lift.
Is FAQ schema worth it for a small business with few pages?
Yes, arguably more so. With few pages, every citation matters and the work is small. Add a real FAQ with schema to your handful of key pages and you have covered your whole site. It is one of the fastest, cheapest ways for a small Kiwi business to raise its odds of being quoted by AI.


