Google officially ended FAQ rich results on May 7, 2026, but the markup itself did not die — and that distinction matters more than the headlines suggest. The value moved off the SERP and into a less visible layer: AI citations. Answer engines like ChatGPT, Perplexity, Google AI Mode, and Copilot still parse FAQPage markup to pull pre-structured question-answer pairs into their responses.
In our experience working with clients across the migration, sites that ripped out the FAQ schema in the first week of May saw their AI citation share decline before any organic ranking impact showed up. The infrastructure they removed was the same infrastructure AI engines were using to attribute them. This piece walks through what actually changed, what didn’t, and what the schema stack needs to look like now.
Key takeaways:
- FAQ schema isn’t dead — Google removed the rich result display, not the markup. Google still uses it for understanding; Bing still renders it.
FAQPageis now a citation layer: schema-rich pages earn higher AI citation rates, though schema alone isn’t enough without authority signals.- Entity verification beats link authority:
Organization/Personschema withsameAsidentifiers lets AI systems resolve and cite you reliably. - Named authors get cited ~1.9× more: credentialed bylines outperform brand-only or anonymous attribution in AI citation research (2026).
- Superficial date updates now trigger penalties:
dateModifiedmust reflect real content changes; Google’s Dec 2025 and Mar 2026 core updates both confirmed this.
What Did Google Actually Change With FAQ Rich Results?
Google stopped displaying FAQ rich results in search on May 7, 2026, completing a phase-out that began in August 2023. Per Google’s own documentation, the FAQPage schema itself remains a valid Schema.org type, and Google confirmed it still uses the markup to understand page content.
The timeline makes the intent clear:
In August 2023, Google restricted FAQ rich results to well-known authoritative government and health websites. The restriction was a response to widespread abuse: sites added artificial FAQ sections to inflate SERP real estate with questions that did not match user intent.
On May 7, 2026, Google removed FAQ rich results for all remaining sites. Google will remove Rich Results Test support in June 2026 and Search Console API support in August 2026 (Search Engine Land, May 8, 2026).
Schema and rich results are not the same thing. Google deprecated the display. It did not deprecate the markup. Sites do not need to remove existing FAQPage structured data.
This is a Google-only move. As of May 2026, Bing still supports and displays FAQ rich results, and other search engines have not announced equivalent changes. FAQPage schema retains potential rich result value beyond Google, on top of its citation value across AI platforms.
Why Does FAQPage Schema Still Matter for AI Search?
Because FAQPage schema is now a primary extraction layer for AI answer engines. FAQ schema provides pre-structured question-answer pairs that AI systems can lift directly, without parsing unstructured prose. Each platform uses that structure differently.
Platform-specific behaviour patterns we’ve observed:
- ChatGPT favours neutral, encyclopedia-style answers with specific data, and uses FAQPage to anchor question-answer mappings.
- Perplexity favours conversational, experience-driven content, and pairs FAQPage with Person schema for attribution.
- Google AI Overviews favours passage-level validated answers, and pairs FAQPage with Speakable for extraction.
The evidence on schema’s exact citation lift is contested. The Searchless Journal reports that pages with thorough schema markup gain roughly a 36% advantage in AI summaries and citations (Searchless Journal, April 2026). Frase.io’s FAQ-schema analysis reports similar advantages for schema-rich pages. But Ahrefs ran a controlled experiment with 1,885 pages that added JSON-LD between August 2025 and March 2026 and found a statistically significant 4.6% decrease in Google AI Overview citations after schema was added (Search Engine Journal coverage).
Our read: schema markup is almost certainly a positive factor in combination with strong content, but it is not a magic bullet. Ahrefs’ result is consistent with Google’s own published guidance — schema “isn’t a load-bearing factor for AI search” on its own. The pages that earn citations have schema and depth, and entity verification, and recent updates.
The Atomic Answer rule makes FAQ schema work harder regardless of which study you weight. The first sentence under each H2 should be a 40–60-word self-contained answer. That is the passage AI systems lift for direct citation. FAQ schema structures extractability at the markup level; the Atomic Answer structures it at the content level. Both together is the durable pattern.
How Does Entity Verification Replace Traditional Authority?
Authority in AI search is increasingly defined by recognized entity status, not raw Domain Authority. This isn’t yet supported by a single published correlation coefficient we can hand you with full confidence — claims of an “r = 0.81” correlation between E-E-A-T signals and AI citations circulate in 2026 commentary, but we have not located the original methodology or peer-reviewed publication, so we are not citing the specific number. What is well-documented is the directional pattern: pages whose underlying entities are resolved in Google’s Knowledge Graph are cited more often than pages whose authors and publishers are unknown to the system.
Stackmatix’s Organization Schema guide makes the case clearly: the sameAs property is the highest-leverage signal for Knowledge Graph inclusion. It connects your entity to authoritative external profiles Google already trusts.
The sameAs Priority List
To resolve your brand as a discrete entity, your Organization Schema must link to:
- Wikidata: The primary input for Google’s Knowledge Graph.
- LinkedIn Company Pages: Professional existence validation.
- Crunchbase: Business structure and funding confirmation.
- Government Registrations: Official legitimacy.
Person Schema and Named Author Attribution
Anonymous content is a dead-end pattern for AEO. Per research summarized by Am I Cited and corroborated across multiple 2026 analyses, content with clear author attribution earns roughly 1.9x more AI citations than brand-only content, with credentialed authors lifting that to roughly 2.3x.
Person Schema Minimums
- name matching the byline exactly.
- jobTitle reflecting a specific role.
- worksFor linked to the Organization entity.
- sameAs array linking to LinkedIn, Google Scholar, or industry profiles.
Additional properties strengthen the signal: hasCredential for certifications, knowsAbout for topic expertise, alumniOf for educational background. The sameAs array is the most critical property on both Organization and Person schema. Without it, AI systems cannot verify who you are.
What Schema Stack Do You Need for AEO-Ready Brand Authority?
AEO-ready brand authority requires two schema layers: an entity foundation (Organization, Person, sameAs) and a content layer (FAQPage, Article, Speakable). The entity foundation tells AI systems who you are. The content layer tells them what to cite.
- Organization schema handles brand entity declaration for entity resolution and trust scoring.
- Person schema handles author identity verification as an E-E-A-T signal with a 1.9x citation lift.
- FAQPage schema serves as a Q&A extraction layer for direct answer extraction by AI.
- Article/BlogPosting schema handles content classification for freshness signals and content type.
- Speakable schema serves as a passage-level citation target for precision extraction in AI answers.
- sameAs on Organization and Person handles cross-platform entity verification for Knowledge Graph inclusion.
Step-by-Step Implementation Strategy for AEO:
Step 1: Establish the trust layer (identity).
Implement the Organization and Person schema as your foundation. Include a complete sameAs property linking your brand and authors to authoritative external profiles. This verifies your entity for the Knowledge Graph. We’ve watched this single step — adding Wikidata and LinkedIn to a previously unverified Organization schema — move the needle on Knowledge Panel accuracy within weeks for clients who had been invisible to AI Mode for months.
Step 2: Define the content layer (structure).
Add specific schema types so AI engines can extract: FAQPage for direct Q&A extraction; Article for freshness and dateModified signals.
Step 3: Optimize for conversational search (headings).
Write H2 headings as conversational, long-tail questions that mirror how users phrase queries aloud.
Step 4: Deploy the citation signal (Speakable).
Apply Speakable markup to the two-to-three-sentence answer passage immediately following each H2. This tags the content AI systems and voice assistants should treat as the definitive answer.
Step 5: Achieve voice search synergy.
Matching a spoken-style H2 with citable Speakable markup positions the page as the default result for voice queries across Siri, Alexa, and Gemini Live.
How Does Content Freshness Affect AI Citation Trust?
AI systems use dateModified to categorize content into citation weight classes:
- Fresh (0–30 days): high citation probability
- Recent (30–180 days): standard weight
- Aging (180 days–2 years): weighted down
- Stale (2+ years): often excluded from AI results
Integrity over faking. Updates must be substantive. Google’s December 2025 core update specifically targets superficial date changes without content improvements (Dataslayer recovery guide); the March 2026 core update continued and surpassed that trajectory in volatility. Updating dateModified without real changes risks a domain-wide date-weight discount.
Avoid schema drift. Ensure markup and content align. If your schema claims a recent update but the text references outdated data, AI systems detect the inconsistency — a pattern increasingly referred to as schema drift — and trust degrades fast.
Why Does Topical Authority Matter More Than Individual Rankings?
AI citation systems favour depth over breadth. A domain with one viral post on a topic rarely gets cited. A domain with thirty connected pages on the same topic gets cited consistently across dozens of related queries. Search Engine Journal’s coverage of the post-Gemini-3 landscape echoes this: brands are moving past fragmented keyword rankings and toward recognized “entity” status inside the AI Knowledge Graph.
When ChatGPT, Perplexity, or Google AI Overviews need to cite a source, they draw from sources their retrieval systems have come to recognize as consistently accurate on that subject. Deep, structured, frequently-updated content becomes the default reference. Surface-level content earns nothing. In our experience, the threshold where a topic cluster starts compounding sits around 20–30 connected pages with shared entity schema — below that, citations are sporadic; above it, AI engines start treating the domain as a default reference for the entire topic.
Entity integration connects topical authority to the schema stack. Brands need to transform from keyword collections into recognized concepts inside the AI Knowledge Graph. That requires schema linking entity relationships between pages, internal linking that surfaces cluster relationships, and backlinks using entity-relevant anchor text.
Topical clusters are the content strategy. Entity schema is the technical verification layer. A deep content cluster without entity schema is harder for AI systems to attribute. Entity schema without deep content gives AI a verified identity but nothing substantive to cite.
Need Help? Contact Us
The shift from SERP features to entity authority is the new operating model for search visibility. If your team is ready to build the schema infrastructure that makes AI engines cite you by name, reach out to Blacksmith SEO and we will map the implementation to your site.
Frequently Asked Questions About FAQPage Schema and Brand Authority
Should I Remove FAQ Schema Now That Google Dropped Rich Results?
No. Google confirmed it still uses FAQPage structured data to understand pages, Bing still displays FAQ rich results, and AI answer engines use FAQ schema as an extraction layer for citations. Removing it eliminates a verified citation pathway across multiple platforms while saving you nothing meaningful in technical debt.
Does FAQPage Schema Help With AI Overviews?
FAQPage schema provides AI systems with pre-structured question-answer pairs they can extract directly. Several 2026 studies (Frase.io, the Searchless Journal) suggest schema-rich pages earn higher AI citation rates, though Ahrefs’ controlled experiment found schema alone did not lift citations in their dataset. The honest answer: schema is a meaningful piece of a larger pattern, not a standalone lever.
What Is the Most Important Schema Type for Brand Authority?
Organization schema with a complete sameAs array is the foundation. It declares your brand as a verified entity and connects it to authoritative external profiles Google already trusts. Without Organization schema, AI systems struggle to resolve your brand identity against Knowledge Graph records, which reduces trust scores in citation decisions.
Does Having a Named Author Improve AI Citations?
Yes. Research summarised across multiple 2026 analyses found content with clear author attribution earned roughly 1.9x more citations from AI systems than anonymous or brand-only content, with credentialed authors performing even better. Person schema with name, jobTitle, worksFor, and sameAs properties creates the verifiable identity AI engines use to evaluate expertise.
How Often Should I Update dateModified for Freshness Signals?
Only when you make substantive content changes. AI systems treat recently modified content as fresh and weight it up, but Google’s December 2025 and March 2026 core updates penalize superficial date changes. Every refresh should add new data, update outdated references, or expand topic coverage — otherwise the freshness signal gets discounted across your entire site.
About this guide
Primary sources: Google Search Central’s FAQ structured data deprecation notice (posted May 7, 2026); Barry Schwartz, Google to no longer support FAQ rich results, Search Engine Land (May 8, 2026).
Secondary sources: SE Ranking’s Gemini 3 impact study (100,000 keywords, Feb 2026); Ahrefs’ AI citation freshness study (16.975 million citations, July 2025); Stackmatix’s Organization Schema guide on sameAs and Knowledge Graph; and the Searchless Journal on schema as a citation layer. Numbers attributed to studies are quoted from the source publication; reach out if you’d like the underlying CSVs or methodology notes.
