An faq schema seo guide serves one clear purpose: telling you exactly what FAQ structured data does for your search visibility in 2026, whether it is still worth implementing, and how to do it correctly. Google retired the expandable accordion rich results that FAQ markup used to trigger — but the structured data itself now carries real weight as a signal for SEO in South Africa, particularly for AI-powered search answers.
This guide covers the full picture: what changed, what survived, and the implementation steps that take under ten minutes per page.
The timing matters. Google confirmed the FAQ rich result deprecation on May 7, 2026, leaving many South African businesses unsure whether to remove their FAQ markup or leave it in place. Most published advice is still catching up. This guide draws on Google's own structured data documentation and post-deprecation tracking data so you are not acting on 2022-era guides.
If you have already read our overview of schema markup for South African websites, this post goes deeper on the FAQ-specific type and its 2026 status.
Quick Answer
An faq schema seo guide covers structured data markup that tells search engines a page contains questions and answers in a machine-readable format. Google deprecated FAQ accordion rich results on May 7, 2026 — the expandable dropdown no longer appears in search results. However, the FAQPage schema type remains valid and actively feeds Google AI Overviews, Perplexity, and other AI search tools. Correct implementation is more strategically valuable now than it was before the deprecation.
In This Guide
Not Sure Which Schema Types Your Site Is Missing?
Send us your URL and we will run a structured data audit — showing you exactly which schema gaps are costing you AI search citations and rich result eligibility.
Get a Free Schema AuditWhat Is FAQ Schema?
FAQ schema is a type of structured data designed to mark up a list of questions and their corresponding answers, using the FAQPage type defined by schema.org. It signals to search engines that specific sections of your page are genuine question-and-answer pairs — not inferred from context, but explicitly labelled in machine-readable code.
Structured data uses a standardised vocabulary that Google, Bing, and other search engines read to understand content beyond raw text. When you wrap your FAQ section in FAQPage markup, you are attaching a label that reads: "These specific texts are questions, and these other specific texts are their complete answers." That explicit mapping lets search engines use your content in AI answers, featured snippets, and other search features without guessing at meaning from surrounding prose.
If you are new to the concept, our guide on what schema markup is and how it works in South Africa covers the broader vocabulary before you dive into a specific type.
Google's preferred format is JSON-LD — a JavaScript block in the page's <head> or <body>, completely separate from the visible HTML. It can be injected dynamically without touching the page layout. The other supported formats are Microdata and RDFa, but both require attributes embedded inside your HTML elements — harder to maintain as content changes.
Schema.org vs. JSON-LD: Schema.org is the vocabulary — the list of types and properties. JSON-LD is the syntax — how you write it. FAQPage schema in JSON-LD uses the FAQPage, Question, and Answer types from schema.org. Both are required; neither works without the other.
What Google Changed in May 2026
On May 7, 2026, Google removed FAQ rich results from Google Search entirely — the expandable question-and-answer accordions that used to appear below some listings are no longer shown for any site or any search query.
The full timeline:
- August 2023: Google restricted FAQ rich results to government and health authority sites only. Most commercial sites lost eligibility here.
- May 8, 2025: Google added a formal deprecation notice to its FAQ structured data documentation, giving twelve months' notice of full removal.
- May 7, 2026: FAQ rich results stopped appearing in Google Search completely.
- June 2026: Search Console FAQ reporting and testing tools removed.
- August 2026: API support discontinued.
What this means practically: if you have FAQ schema on your site right now, the markup is harmless and still valid per schema.org. Google confirmed there is no penalty for keeping it. What will not happen is the old accordion dropdown appearing below your search listing — that visual feature is gone.
The mistake to avoid is conflating the two separate things: the rich result feature (gone) and the FAQPage schema type (still valid). Much of the panicked advice to "remove all FAQ schema" in mid-2026 confused them. Removing markup that now serves AI search signals because a visual SERP feature disappeared is the wrong response.
Key Takeaway
The FAQ rich result — the clickable dropdown in search results — is deprecated and gone. The FAQ structured data type is not deprecated. These are two separate things. Keeping well-implemented FAQ markup on pages with genuine Q&A content remains the right call; removing it to "clean up" after the deprecation discards the AI search signal value without recovering anything.
Why FAQ Schema Still Matters for SEO in 2026
FAQ schema now matters for a fundamentally different reason than it did in 2022: it feeds AI-powered search answers rather than visual rich results.
Google AI Overviews, Perplexity, and ChatGPT Search are all doing the same thing when they compile an answer: extracting content from sources they can verify and cite with confidence. A FAQPage markup block hands these systems a pre-matched question-to-answer structure that reduces the interpretive work the AI needs to do. Marked-up content is easier to extract, easier to cite, and less likely to be misrepresented.
Research cited by Frase.io puts pages with FAQ markup at 3.2x more likely to appear in Google AI Overviews compared to pages with equivalent content but no structured data. A separate tracked study by Gatilab found FAQ-marked pages earned 2.1x the AI citation volume on Perplexity and ChatGPT over a 90-day period compared to unmarked equivalents. These are third-party figures, not Google's own data — but the underlying logic is well-supported: question-to-answer mapping mirrors exactly how AI systems present information to users.
The competitive gap matters here too. Only 12.4% of websites implement structured data of any kind, according to research cited by Frase.io. For South African businesses that implement FAQ schema correctly now, the window is open: most local competitors have either removed their markup entirely or never added it.
Following an up-to-date faq schema seo guide — one that reflects what changed in May 2026 and what did not — is what separates implementations that serve AI search from those built on outdated assumptions.
FAQ markup also continues to influence featured snippets — the "position zero" boxes above organic results in traditional search. That value was never coming from the JSON-LD alone; it was always driven by the visible text. Direct, contained answers under question headings earn featured snippets regardless of markup. Structured data adds the explicit label that helps search engines find and extract those answers reliably. If you are working through your on-page SEO checklist for South African sites, FAQ schema deserves a line item.
Key Takeaway
The strategic value of FAQ schema shifted, not disappeared. It moved from "earn a visual dropdown in search results" to "feed AI Overviews and citation engines with pre-structured question-answer data." The implementation effort is identical; the reward is bigger for AI search than the old rich result ever was for a commercial site.
Unsure Whether Your FAQ Schema Is Correctly Implemented?
Share your URL and we will check your JSON-LD syntax, schema-to-content match, and AI citation readiness — and tell you exactly what to fix.
Check My FAQ SchemaYour FAQ Schema SEO Guide to Implementation in Four Steps
FAQ schema implementation takes under ten minutes per page when you follow four steps in the right order.
Step 1: Identify the Right Questions
Extract 5–8 questions that directly address what users on this specific page are already asking. Good sources: Google's "People Also Ask" box for your target keyword, your Google Search Console query data filtered to that URL, and forum discussions (Reddit, Quora, or local platforms where your audience congregates).
Avoid generic questions that could appear on any page — each should be specific to this page's content and reflect how real people search. For a deeper approach to identifying the right questions, the SEO content brief guide covers search intent research in full.
Step 2: Write Direct, Standalone Answers
Each answer should be between 40 and 150 words, leading immediately with the direct response — no warm-up sentence, no "great question." The answer must be fully understandable without context from elsewhere on the page. AI systems and search engines extract answers without surrounding paragraphs. "As mentioned above" is useless when the answer is the only thing being read.
Step 3: Add the JSON-LD Block
If you are using WordPress with Rank Math — the most widely used SEO plugin on South African WordPress sites — there is a built-in FAQ block in the Gutenberg block editor that generates the JSON-LD automatically without any coding. Search for "FAQ by Rank Math" in the block inserter, add it to the page, fill in your questions and answers, and publish. The structured data is written for you.
For manual implementation, the core structure looks like this:
Valid JSON-LD structure:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "Does FAQ schema still work for SEO in 2026?",
"acceptedAnswer": {
"@type": "Answer",
"text": "FAQ schema no longer generates rich result dropdowns in Google Search, but FAQPage structured data remains valid and feeds AI Overviews and other AI search tools. It is worth implementing correctly on pages with genuine question-and-answer content."
}
},
{
"@type": "Question",
"name": "What format does Google recommend for structured data?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Google recommends JSON-LD as the primary format for structured data. It is separate from the page's HTML, easier to maintain, and can be injected dynamically without altering the visible content."
}
}
]
}
</script>Step 4: Validate
Run the page URL through Google's Rich Results Test (search.google.com/test/rich-results). Even though FAQ no longer produces a rich result, the tool validates JSON-LD syntax and confirms that schema content matches your visible page text — a requirement Google enforces. Revalidate any time you edit the FAQ section or the structured data.
What Makes Good FAQ Content for Structured Data
The markup signals the structure; the content determines whether the answer gets cited.
Google's structured data policy is explicit: schema markup must accurately describe visible page content. You cannot mark up content that users cannot see. This means your FAQ section must exist as readable on-page content, and your JSON-LD must reflect that content accurately. Any mismatch fails validation and can draw a manual action.
For AI search citation purposes, these qualities separate cited content from skipped content:
- Specificity: "FAQ rich results were deprecated on May 7, 2026" gets cited. "FAQ schema changed recently" gives AI systems nothing clean to extract.
- Completeness: Each answer should make full sense without the reader scrolling the rest of the page.
- Freshness: Outdated answers create conflicting signals when search engines compare your markup against newer sources. Build FAQ updates into your content maintenance schedule.
- Conversational phrasing: Write questions the way a person types them. "How do I add FAQ schema to WordPress?" reflects genuine intent more closely than "FAQ schema WordPress implementation guide."
Common FAQ Schema Mistakes and How to Fix Them
Four implementation errors account for the majority of FAQ schema failures, and all four are fixable in under five minutes once you know what to look for.
Curly quotes instead of straight quotes. Word processors and some CMS editors auto-format straight quotes (" ") into curly/smart quotes (" "). JSON-LD requires straight quotes. Curly quotes break the syntax silently — the code looks correct to the eye but fails the validator. Fix: paste your JSON-LD through a plain text editor before adding it to your site, or use your plugin's built-in fields to avoid writing raw JSON.
Schema content mismatched with visible page text. If the JSON-LD answer says one thing but the on-page FAQ section says something different — or is missing from the page entirely — Google flags it as a policy violation. Fix: write the visible FAQ section first, then copy the text into your JSON-LD. This order also helps because visible content quality is what drives the citation; the markup is the label.
Too many questions. Stuffing 20-plus questions into FAQ schema dilutes quality and creates a maintenance burden. The working range is 5–8 questions per page, each directly relevant to that page's topic. Fix: prioritise questions with evidence of real search demand, cut anything that is "nice to have," and add more only when you can source the questions from actual search data.
Duplicate schema from multiple plugins. Running Rank Math and Yoast simultaneously, or enabling FAQ schema in both a plugin and a page builder like Elementor, generates conflicting markup that Google's Rich Results Test will flag. Fix: designate one tool per page for schema output and disable schema generation in all others for that page. If you are auditing an existing site for these issues, the technical SEO checklist for South African websites has a structured data section.
Want a Full Technical SEO Review?
Share your site and we will assess your structured data implementation, indexation health, and on-page signal gaps — then tell you which fixes will move the needle first.
Book a Technical SEO ReviewWhy South African Businesses Choose Growth Pulse Media for Technical SEO
Dirk built and scaled a South African ecommerce operation before founding Growth Pulse Media. That background means the SEO work here is done by someone who has tracked schema implementation through real revenue impact — not someone pattern-matching from a checklist. When Google retired FAQ rich results in May 2026, the decision on whether to keep or remove structured data on each client's pages was made from first principles, not from a vendor memo.
GPM works with a deliberately limited client load — senior attention on every account. Schema decisions after the 2026 change, prioritisation across South African SEO strategy, and AI search positioning are made by the operator running the campaign, not delegated down. If your site has structured data gaps or is invisible to AI search tools, the work starts with what is actually on your pages.
Who This Is NOT For
Not for businesses manufacturing FAQ content purely for markup purposes. If the plan is to write keyword-stuffed "questions" that the page does not genuinely answer, this is the wrong approach. Google's structured data policy requires that markup reflects real, visible page content. Fabricated FAQ sections built to exploit schema fail validation and create manual action risk.
Not for sites expecting the old accordion rich result. If the objective is the visual dropdown that used to appear below search listings, that feature was retired in May 2026 and will not return. If accordion UI is what you need for UX purposes, it needs to be built into the page itself — schema cannot trigger a display feature that Google has removed.
Not for businesses unwilling to maintain the schema. Outdated answers create conflicting signals. If you implement FAQ markup and never update it as policies change, prices change, or answers become stale, the structured data becomes a liability rather than an asset. Schema requires the same maintenance schedule as the page content it describes.
Not for thin pages expecting schema to compensate for shallow content. A six-question FAQ bolted onto a 300-word page will not earn AI Overview citations regardless of how clean the JSON-LD is. FAQ schema amplifies good content; it cannot substitute for it. If your pages need a depth audit before schema becomes useful, understanding how to prioritise SEO fixes is a better first step.
The questions below round out this faq schema seo guide with the implementation decisions South African businesses raise most often.
Frequently Asked Questions About FAQ Schema
Does FAQ schema still work for SEO after the May 2026 deprecation?
FAQ schema no longer generates the expandable accordion in Google Search results — that display feature was retired on May 7, 2026. However, the FAQPage schema type remains valid, is still read by Google and other AI search engines, and contributes to AI Overview citations and featured snippet eligibility. The use case shifted from visual rich results to AI search signals, but the implementation method and the content requirements are largely the same.
How many FAQ questions should I include in my schema markup?
Between five and eight questions per page is the research-backed working range. Fewer than five gives the schema limited depth; more than eight tends to dilute quality and increase the maintenance burden. Each question should be specific to that page's topic and reflect an actual search query — People Also Ask data and Google Search Console query reports are the two best sources for identifying which questions to use.
Do I need to know how to code to add FAQ schema to my website?
Not if you are using WordPress with Rank Math or Yoast SEO. Both plugins include built-in FAQ blocks in the Gutenberg editor that generate the JSON-LD automatically. You type the questions and answers as plain text; the plugin handles the syntax. For custom-built sites, the JSON-LD structure is straightforward to write manually — the critical requirements are straight quotes (not curly), matching schema content to visible page text exactly, and validating through Google's Rich Results Test before publishing.
Can I keep existing FAQ schema on pages that do not have a dedicated FAQ section?
Google's policy requires that schema markup accurately describes visible page content — if the questions and answers exist somewhere on the page, the markup is valid. If the FAQ section was removed from the visible page but the JSON-LD remains, the mismatch creates a policy violation. Audit your existing FAQ schema implementations to confirm the visible content and the structured data still match; where they do not, either restore the visible section or remove the markup.
What is the difference between FAQ schema and HowTo schema?
FAQ schema (FAQPage) marks up a list of questions and their answers — it is appropriate for reference content where users ask distinct questions that each have a self-contained answer. HowTo schema marks up a sequential process with steps — it is appropriate for instructional content where the order of steps matters. For a page that both explains a process and answers related questions, you can implement both schema types on the same page as long as the content supports both.
Get Your Schema Implementation Right the First Time
Growth Pulse Media audits structured data for South African websites — checking JSON-LD syntax, schema-to-content alignment, AI citation readiness, and which schema types your site is missing entirely. Senior attention on every account. Named platforms, real technical work.
No obligation — we will get back to you within 24 hours.
Talk to a South African SEO Specialist