FAQ schema (formally FAQPage in schema.org) is structured data added to a webpage's source code that explicitly labels question-and-answer pairs. It tells search engines and AI engines that a section of the page contains FAQ content, which makes the content easier to parse, cite, and surface in AI Overviews and similar features.

FAQ Schema for Local Businesses: The 20% Citation Lift

Q: How much does FAQ schema actually help?

Recent studies through 2026 show pages with FAQPage schema earn citations from ChatGPT, Gemini, Perplexity, and Google AI Overviews at roughly 20% higher rates than pages without it, holding content quality constant. The exact number varies by query type and competition. The effect is largest for question-style queries common in local search.

Q: Does FAQ schema still help in traditional Google search?

Yes, indirectly. Google removed the FAQ rich-result feature from most desktop search results in August 2023, but FAQPage schema still helps AI Overviews comprehension, voice search responses, and Google Business Profile question matching. It is no longer a SERP rich-snippet play; it is an AI-search and structured-data play.

Q: Where should FAQ schema go on a local business site?

Place it on the homepage (general business questions), every primary service page (service-specific questions), and the most-trafficked blog posts. The schema must reflect actual visible content on the page; Google penalizes hidden or misleading FAQ schema. Three to seven Q&A pairs per page is the sweet spot.

The short answer

FAQ schema (formally FAQPage JSON-LD) labels question-and-answer content on a webpage so search engines and AI engines can parse it cleanly. Pages with it earn citations from ChatGPT, Gemini, Perplexity, and Google AI Overviews at roughly 20% higher rates than equivalent pages without it. For local businesses, it is the single highest-ROI technical change available.

What FAQ schema actually is

Schema.org is a shared vocabulary that search engines and AI engines use to read structured content on webpages. FAQPage is one specific schema type, designed for question-and-answer content. It is added to the page as a JSON-LD script in the <head> section. The visible content on the page does not change. The schema simply tells the engines: "this section is a Q&A block, and here is exactly which words are the questions and which are the answers."

Why this matters for AI search: ChatGPT, Gemini, Perplexity, and Google AI Overviews crawl webpages and try to extract direct answers. Plain HTML paragraphs work, but the engines have to guess at structure. JSON-LD removes the guessing. The engines pull the Q&A pairs directly and cite the page when one of those questions matches a user query.

The 20% number

Multiple AI search studies through 2026 (Panstag, Botify, Search Engine Land coverage of major SaaS analyses) have measured the lift from FAQPage schema on citation probability. The consensus range is 15 to 25%, with 20% as the typical mid-point. The effect is largest for question-style queries ("how much does X cost," "do you need X for Y," "how long does X take") which is exactly the query shape local-service customers use.

The studies hold content quality constant. Two equivalent pages, one with FAQPage schema and one without, see different citation rates from AI engines. The schema is not making the content better; it is making the content easier to parse.

The base template

Here is the FAQPage JSON-LD template every local business can use. Replace the question and answer fields with the actual content from the page. Place inside the <head> section of the HTML.

FAQPage JSON-LD (base template)

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "REPLACE WITH FIRST QUESTION",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "REPLACE WITH FULL ANSWER. Keep under 300 words. Use plain language. Include specifics."
      }
    },
    {
      "@type": "Question",
      "name": "REPLACE WITH SECOND QUESTION",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "REPLACE WITH FULL ANSWER."
      }
    }
  ]
}
</script>

That is the entire pattern. Three to seven Question objects per page is the sweet spot; fewer than three and the engines have too little to work with, more than seven and the schema gets diluted.

Three FAQ templates by category

The questions that matter for AI search are the questions customers actually ask. Below are three starter sets for the most common local-business categories. Adapt the wording, keep the structure.

Home services (HVAC, plumbing, electrical)

"How much does [service] cost in [city]?"
"Do you offer 24/7 emergency [service]?"
"How quickly can you get to [city] for a service call?"
"Are you licensed and insured?"
"How long does a typical [service] take?"

Dental, medical, wellness practitioners

"Do you accept [insurance plan]?"
"How long is a typical first appointment?"
"Do you see new patients?"
"Where are you located and what are your hours?"
"What should I expect at my first visit?"

Legal services

"Do you offer free consultations?"
"What types of [legal area] cases do you handle?"
"How long do [case type] cases typically take?"
"What does it cost to hire a [practice area] attorney?"
"Do you work on contingency or hourly?"

The pattern across categories: cost, timing, location, eligibility, expectations. These are the five question shapes customers ask before booking. The same five shapes apply to almost every local-service category.

Quick winOpen a notebook. Write down the 5 questions you get most often by phone or email before a customer books. Those exact 5 questions, with your real answers, become your FAQPage schema. Total writing time: 30 minutes. Total deployment time: 10 minutes if your site uses WordPress with a schema plugin.

Want us to ship the FAQ schema for your site?

Free 30-minute Strategy Session. We will look at your top 5 customer questions, write the FAQPage JSON-LD, and show you exactly where it goes.

Book a Strategy Session →

Where to place FAQ schema on a local site

Not every page needs schema, but the high-leverage placements are:

Homepage. 3 to 5 general business questions (hours, location, services, insurance, booking process).
Every primary service page. 4 to 7 questions specific to that service (cost, timing, what is included, common follow-ups).
Top blog posts. 3 to 5 questions related to the article topic, mirroring the article's own Q&A section if it has one.
Pricing page if you have one. 3 to 5 questions about pricing, payment, and what is included.

The schema must reflect actual visible content on the page. Google's structured-data guidelines (and AI engines' parsing logic) penalize pages where the schema does not match the visible page. The simplest safe pattern: put the schema in the <head>, then put the same questions and answers as visible text in an FAQ section on the page itself.

What to avoid

Three failure modes that kill the lift:

Hidden FAQ content

Schema for questions and answers that are not visible on the page. The schema must mirror visible content. Hidden schema can trigger structured-data warnings in Google Search Console and reduces AI citation reliability.

Generic AI-generated questions

"What are the benefits of [service]?" is a generic prompt; real customers do not ask in that voice. The schema should mirror the actual phrasing customers use on the phone. Specific is better than polished.

The same FAQ block on every page

Duplicating identical FAQ schema across the site dilutes the value. Each page should have questions specific to what that page is about. Homepage questions are different from service-page questions are different from blog-post questions.

How long until you see results

Realistic timeline: AI engines typically reindex schema changes within 1 to 4 weeks. Citation lift from new schema becomes measurable around the 30 to 60-day mark for most local categories. Larger lifts (the upper end of the 20% range) require the rest of the AEO foundation to be in place too: a Person schema with sameAs chain, EEAT signals like author bios and dateModified, and internal linking.

FAQPage schema alone is the floor. It produces some lift on its own. It produces more lift as part of the full AEO play described in our AEO guide for local businesses.

The bigger picture

AI search visibility is becoming the channel local businesses underweight at their peril. 45% of local-service consumers now use AI search to find businesses, up from roughly 6% twelve months ago (SOCi 2026 local AI study). Most local businesses have zero schema on their site, which means AI engines have to guess at their structure, which means they often do not get cited. The fix is mechanical and inexpensive. FAQPage schema is the first move.

If you want the full AEO foundation built and maintained, paired with the rest of the marketing stack, that is what our AI Search service covers under a single retainer.

Frequently asked questions

What is FAQ schema?

FAQ schema is structured data added to a webpage's source code that labels question-and-answer content. Search engines and AI engines use it to parse, cite, and surface FAQ content reliably.

How much does FAQ schema actually help?

Pages with FAQPage schema earn citations from ChatGPT, Gemini, Perplexity, and Google AI Overviews at roughly 20% higher rates than pages without it, holding content quality constant.

Does FAQ schema still help in traditional Google search?

Yes, indirectly. Google removed FAQ rich-result snippets from most SERPs in 2023, but FAQPage schema still feeds AI Overviews, voice search, and Google Business Profile question matching.

Where should FAQ schema go on a local business site?

Homepage, every primary service page, top blog posts, and the pricing page. Three to seven Q&A pairs per page is the sweet spot. The schema must reflect actual visible content on the page.