Schema & Structured Data
27 practices
29.6% of websites still have zero JSON-LD schema in 2026 (Foglift AI Readiness Study)
Foglift Q2 2026 study of 1,386 scans across 344 domains: 29.6% have zero JSON-LD. Median AI search subscore: 46/100 vs 86/100 median SEO readiness. Huge opportunity gap.
Sources
29.6% of websites still have zero JSON-LD schema in 2026 (Foglift AI Readiness Study)
Foglift Q2 2026 study of 1,386 scans across 344 domains: 29.6% have zero JSON-LD. Median AI search subscore: 46/100 vs 86/100 median SEO readiness. Huge opportunity gap.
Sources
71% of ChatGPT-Cited Pages Include Structured Data; Product+AggregateRating Schema = 3x Citations
Schema markup is a stronger AI citation signal than previously understood, with specific schema types driving disproportionate visibility gains.
Sources
Attribute-Rich Schema Outperforms Generic Schema by 20pp for AI Citations
Generic schema provides zero citation advantage (OR=0.678, p=.296). Product/Review schema with populated pricing/ratings: 61.7% vs 41.6%.
Sources
Comparison tables, llm.txt, and FAQ schema lift AI visibility 28-34%
Erlin's controlled test across 500+ brands: comparison tables +34%, llm.txt +32%, FAQ schema +28% coverage lift within 14-21 days across ChatGPT, Perplexity, Gemini, Claude.
Sources
Comprehensive schema triples AI citation likelihood, but generic boilerplate schema lowers it to 41.6%
Analysis of 500+ sites found comprehensive schema ~3x more likely to earn AI citations. Generic boilerplate schema has 41.6% citation probability vs 59.8% for no schema.
Sources
FAQPage Schema Still Delivers 28-85% Citation Lift Despite Rich Result Retirement (May 2026)
Multiple 2026 studies confirm FAQPage schema still boosts AI citations after Google retired FAQ rich results (May 7, 2026). Presenc: +55-85% on conversational queries; OpenHermit: 28-40%.
Sources
FAQ Schema markup correlates with 67% higher AI citation rates
WhatsMyGeoScore analysis of 50K+ pages in Orion's GEO Index found FAQ schema delivers the highest citation impact of any schema type for AI-generated answers.
Sources
FAQPage, Article, Organization, HowTo Are Top Schema Types for GEO
FAQPage schema has highest direct citation impact. Organization schema with sameAs property anchors entity graph. Incorrect schema hurts more than no schema.
Sources
FAQPage schema + bullet lists appear in 73% of AI Overviews citations vs 31% for plain paragraphs
BrightEdge 2026: pages with FAQ schema and bullet lists cited 2.35x more. Flesch-Kincaid grade 8-10 cuts dense academic prose inclusion rates by more than half.
Sources
FAQPage Schema Effectiveness Is Contradictory: +41% Citation Lift vs No Effect (Multiple Studies 2026)
Studies conflict: Causabi/Cyberclick report 41% FAQPage citation lift; Ahrefs 1,885-page study found +2.4% AI Mode (insignificant), +2.2% ChatGPT, -4.6% AI Overviews. GrackerAI reports 3.7× lift.
Sources
Gemini Is Most Schema-Responsive Engine — 52% Citations from Brand-Owned Sites, Only 13.7% Overlap with AI Overviews
Yext 6.8M citation study: Gemini cites brand-owned sites at 52% — highest of any engine. Gemini and AI Overviews share only 13.7% of citations on same queries.
Sources
Triple JSON-LD Stack (Article + ItemList + FAQPage) Yields 1.8x More AI Citations
GenOptima's 2026 GEO Playbook found triple JSON-LD schema stack (Article + ItemList + FAQPage) yields 1.8x more citations. List pages optimized for 7-14 day freshness cycles achieve highest citation
Sources
Hyperlocal Schema Stack for AI: LocalBusiness + Service + Review + FAQPage + Payment
AI Search Shift 2026: hyperlocal schema stack — LocalBusiness, Service, Review, FAQPage, OpeningHoursSpecification, acceptedPaymentMethod. Every customer filter needs a structured field.
Sources
Full-Stack JSON-LD (Rich Completeness Tier) Cited at 2-3x vs Thin Schema — Completeness Beats Presence
Toronto SEO 2026 study: pages with Article+FAQPage+Person schema cited at 2-3x rate vs schema-thin equivalents controlling for rank. Perplexity shows largest schema lift of any engine.
Sources
JSON-LD Schema Revival — 65–71% of Cited Pages Carry Structured Data
Pages cited by Google AI Mode (65%) and ChatGPT (71%) carry structured data (SE Ranking 2026). Four schema types non-negotiable: Article, BreadcrumbList, Organization, Person.
Sources
Pages with schema markup have 2.5x higher AI citation rate (Stackmatix 2026)
Structured data boosts AI citations. Content with proper schema sees 2.5x higher chance in AI answers. FAQPage schema improves citation rates by 30% on average.
Sources
Pages with schema markup have 2.5x higher AI citation rate (Stackmatix 2026)
Structured data boosts AI citations. Content with proper schema sees 2.5x higher chance in AI answers. FAQPage schema improves citation rates by 30% on average.
Sources
Schema FAQPage Triggers in 89% of Q&A Queries — 3.2x AI Citation Lift
NeuraPulse testing across 200+ domains: comprehensive schema yields 3.2x AI citation lift. FAQPage triggers in 89% of Q&A queries. Speakable powers 60% of voice answers.
Sources
Schema markup increases AI citations by 3.2x — ClaimReview + Author URL lifts Perplexity 12% to 34%
NetWebMedia tested schema across Perplexity, ChatGPT, Claude, AI Overviews. ClaimReview+Author schema: Perplexity citation rate jumped from 12% to 34% in 6 weeks. BreadcrumbList helps context-aware ci
Sources
Schema Markup Has Near-Zero Independent Effect on AI Citations — Ahrefs 1,885-Page Controlled Study
Ahrefs tracked 1,885 pages adding schema vs 4,000 controls. Result: zero meaningful citation lift on any AI platform. Schema compounds quality but doesn't stand alone.
Sources
Schema + structural clarity yields 40% higher AI visibility (Princeton GEO study)
Princeton KDD-2024 and AlphaSol Labs 2026: pages with structured lists, statistics, quotes and schema achieve 30-40% higher visibility in AI answers; Wellows measured +73% correlation.
Sources
Structured geo data produces 6x citation rate lift for location-based listings
MapAtlas audited 100 listings across 3 engines: prose-only got 12% citation rate, while full structured geo data reached 71%. Geo coordinates alone provide 3.1x lift.
Sources
Triple-Schema Stacking (FAQPage + Article + HowTo) Lifts Citations 1.8× vs Article Alone
OptimizeGEO research shows implementing FAQPage + Article + HowTo schema via JSON-LD @graph stacking produces 1.8× more AI citations than Article schema alone.
Sources
Verified-Review Depth Is the
Idukki's 1,200-prompt study across five engines found verified-buyer review depth the strongest citation predictor. Complete Review JSON-LD gave a 14x citation rate over text-only reviews.
Sources
Yext 17.2M citation study: verified structured data = 54% higher citation rate across all engines
Yext analyzed 17.2M AI citations across ChatGPT, Perplexity, Gemini, Claude. Verified structured data gave 54% higher citation odds. AI visibility depends on retrieval logic, not just content quality.
Sources
Yext 17.2M citation study: verified structured data = 54% higher citation rate across all engines
Yext analyzed 17.2M AI citations across ChatGPT, Perplexity, Gemini, Claude. Verified structured data gave 54% higher citation odds. AI visibility depends on retrieval logic, not just content quality.