Crypto Schema Markup Guide 2026: FinancialProduct, Cryptocurrency and Beyond

The 4 schema types Web3 sites need

Most Web3 SEO advice doesn't differentiate between schema types. The result: teams ship Article schema everywhere because that's the default. Wrong type for the page type kills rankings.

FinancialProduct for any DeFi protocol, lending market, perp DEX or yield platform. Extends the Service hierarchy. Has crypto-relevant properties: feeStructure, provider, termsOfService, audited contract address as identifier.

Cryptocurrency for tokens. Properties: name, symbol, contractAddress, blockchainNetwork, totalSupply, circulatingSupply. Plus PriceSpecification for live price. AI engines actively look for this when answering "what is the price of X token" queries.

CryptoExchange for CEX and DEX. Extends FinancialProduct. Adds currenciesAccepted, paymentAccepted, hasProofOfReserves URL, areaServed for jurisdictions. Less than 15% of exchanges implement this correctly.

Service for cross-chain bridges and chain-level products with serviceType set to a specific category. Use for L2s, bridges, oracle networks. Plus FAQPage for the safety questions.

Plus the universal stack: BreadcrumbList on every page, FAQPage with explicit Q&A blocks (5+ per key page), Article + Person on blog posts, Review on comparison pages, Organization for the publisher entity. These complement the crypto-specific types.

What you don't need: generic Product schema (use Cryptocurrency or FinancialProduct instead), generic Article on protocol pages (use FinancialProduct), generic Organization on token pages (use Cryptocurrency). Specific types beat generic ones for rich results.

FinancialProduct: protocol pages

FinancialProduct is the foundation schema for any DeFi protocol. Most teams use Article. Wrong.

The minimum viable FinancialProduct schema: @type set to FinancialProduct, name (your protocol), description, url, provider (Organization with logo), termsOfService URL, feeStructure (use Offer or PriceSpecification with eligibleQuantity for tier thresholds).

For DEXes specifically: add interestRate (zero for swap-only DEXes), serviceArea (chains supported), audience (intendedAudience entries for retail vs institutional). For lending protocols: add interestRate as the variable APY range, plus collateral requirements as additionalProperty.

The contract address property: use identifier with PropertyValue sub-schema. propertyID set to "contractAddress" or "auditedContract." value set to the 0x address. This makes your contract address machine-readable for AI engines.

Stacking with FAQPage: a single page can emit both FinancialProduct and FAQPage in the same JSON-LD graph. Use @id references to link the entities. AI engines parse the combined graph and extract both rich result types.

The audit firm citation pattern: add a separate Article in the schema graph for each major audit, with author Person sub-schema for the audit firm. This builds the citation graph that AI engines use for E-E-A-T scoring.

What breaks it: nesting Organization inside provider without @id, missing required properties (name, description), using string for feeStructure instead of structured data, missing url. Validate before deploying.

Cryptocurrency: token pages

Token pages get the wrong schema treatment 9 times out of 10. Cryptocurrency schema is purpose-built and most teams skip it.

The full Cryptocurrency schema: @type set to Cryptocurrency. Properties: name (token name), symbol (ticker), contractAddress (the 0x address), blockchainNetwork (Ethereum, Solana, Polygon etc.), totalSupply, circulatingSupply, maxSupply, decimals, currency (the symbol again for currency code).

Live price integration: add PriceSpecification as separate entity in the graph with price, priceCurrency, validFrom timestamp. Update the timestamp on every price refresh. Stale timestamps signal abandonment to both Google and AI engines.

Tokenomics as additionalProperty: use PropertyValue entries for vesting schedule, governance utility, fee accrual, deflationary mechanics. This is undersupplied in current Web3 and creates AEO citation opportunity.

The where-to-buy block: separate Service entries with serviceType set to "Token Listing", provider as the exchange, plus offers describing trading pair availability. Lets AI engines answer "where can I buy X token" queries by extracting this directly.

For governance tokens: add GoverningBody as additionalType with members property listing core team or DAO members. Plus VotingProcess if applicable. Power users searching for governance details want this.

Validation gotchas: Cryptocurrency isn't in Google's primary rich result types yet so Rich Results Test may show warnings. Schema.org Validator confirms syntax. AI engines parse it regardless of Google's formal support.

CryptoExchange: CEX and DEX pages

CryptoExchange schema (extends FinancialProduct extends Service) is the most undersupplied schema type in crypto. Less than 15% of exchanges implement it.

The full schema: @type set to CryptoExchange. Inherits all FinancialProduct properties. Adds: currenciesAccepted (USD, EUR, USDT etc.), paymentAccepted (bank transfer, card, crypto), areaServed (ISO country codes), hasProofOfReserves URL.

For decentralized exchanges: CryptoExchange still applies. Set provider to the DAO. currenciesAccepted lists supported assets. areaServed can be "Global" if no geo restrictions. The schema doesn't care if you're custodial or non-custodial.

Per-pair pages: ExchangeRateSpecification with currentExchangeRate (live), price, priceCurrency, validFrom. Stack with the parent CryptoExchange via @id reference.

Fee tier schema: use Offer entries with eligibleQuantity (volume threshold) and price (fee percentage). Multiple tiers as separate Offers. Plus PriceSpecification for the per-trade fee structure.

The Proof of Reserves angle: hasProofOfReserves should be a URL to your dedicated PoR page. Plus separate Article schema for the most recent attestation with author (audit firm) and dateModified current.

Geo properties matter for AEO: AI engines use areaServed when answering "is X exchange available in my country" queries. Get this wrong and you don't get cited for geo-specific queries.

Custom properties for Web3 specifics

Schema.org doesn't have built-in properties for everything Web3 needs. Use additionalProperty with PropertyValue to extend cleanly.

The pattern: additionalProperty: [{"@type": "PropertyValue", "propertyID": "customName", "value": "the value"}]. Use propertyID as a stable identifier and value for the actual data.

Common Web3 custom properties: totalValueLocked (for DeFi protocols), volume24h (for DEXes), apy (for yield platforms), mintingFee (for NFT marketplaces), royaltyEnforcement (for NFT platforms), bridgeWithdrawalPeriod (for L2 bridges), validatorCount (for chains).

The Wikidata sameAs link: if your project has a Wikipedia or Wikidata entry, link via sameAs. AI engines, especially ChatGPT, use Wikidata for entity disambiguation. Major AEO enable.

Linking related entities: use @id throughout your schema graph and reference entities via {"@id": "https://example.com/entity#id"}. Lets AI engines understand relationships (your protocol is on this chain, has this token, audited by these firms).

The structured data graph: use @graph at the top level wrapping multiple entities. Cleaner than nesting and easier for AI engines to parse. All Crawlux pillar guides use this pattern.

Validation tools and process

Schema deployed without validation breaks silently. Multiple validators each catch different issues. Run all of them.

Schema.org Validator (validator.schema.org): catches syntax errors and undefined properties. Run this first because it's the strictest. Tests pure schema.org compliance regardless of any specific search engine.

Google Rich Results Test (search.google.com/test/rich-results): tests what Google will actually show. Reports warnings and errors specific to Google's rich result eligibility. Some valid schema doesn't qualify for rich results because Google hasn't opted to surface that type yet.

Bing Markup Validator: Bing's parser is stricter than Google's. Important because ChatGPT uses Bing for retrieval. Schema that passes Google may still fail Bing. Run before shipping anything important.

The validation order: Schema.org Validator → Bing Markup Validator → Google Rich Results Test. Pass all three. Fix in that order because syntax errors block everything else.

CI integration: automate validation on every release. Schema-org-validator has an API. Run against staging before deploys. Most teams skip this and deploy broken schema.

The render-time vs build-time question: SSG sites get easier validation (test the static HTML output). SSR/CSR sites need validation against the rendered DOM, which adds complexity. Use Puppeteer to render then extract JSON-LD then validate.

10 common schema errors I see weekly

Same recurring patterns from 200+ Web3 audits. Most are 5-minute fixes once you know they exist.

Error 1: Multiple H1 tags trigger schema confusion. Fix at the HTML level first. Schema can't compensate for broken HTML structure.

Error 2: Article schema on FinancialProduct pages. Most common error in DeFi. Migrate to FinancialProduct.

Error 3: Missing required properties. Schema.org marks some as required (name, description). Validators flag this. Fix before deploy.

Error 4: Wrong @context URL. Should be https://schema.org. Sometimes shows as http://schema.org or schema.org without protocol. Use exact: https://schema.org

Error 5: Date format inconsistencies. datePublished and dateModified should be ISO 8601 (2026-05-03T10:30:00Z). Plain date strings get rejected by some validators.

Error 6: Image objects without dimensions. ImageObject needs width and height set. Without them, Google doesn't use the image for rich results.

Error 7: Author schema as string instead of Person. Use {"@type": "Person", "name": "Author Name"} not just "Author Name." Person sub-schema is required for E-E-A-T.

Error 8: Mixing JSON-LD and Microdata on the same page. Google deduplicates but it's sloppy. Pick one (preferably JSON-LD) and stick with it.

Error 9: FAQPage with too few questions. 1-2 FAQs is below the threshold for FAQPage rich result eligibility. Need 3+ minimum. Aim for 5+.

Error 10: Schema in head instead of body. JSON-LD can go in either but body is preferred. Some screen readers struggle with head-injected schema.

Schema and AEO citation rates

AI engines parse structured data heavily. Schema correctness predicts AEO citation rate better than any other single signal.

The data we have: across 200+ TG3 client audits, sites with proper crypto-specific schema (FinancialProduct, Cryptocurrency, CryptoExchange) get cited by AI engines 3-4x more often than sites with generic Article schema. Schema density predicts citation rate.

Why it works: AI engines (ChatGPT, Perplexity, Claude) parse JSON-LD when retrieving content. Structured data gives them clean entity extraction (this is a protocol, this is its TVL, this is its token). Without schema, they have to extract from prose, which is noisier.

The Speakable schema enable: Speakable wraps key answer paragraphs and signals to AI engines "this is the part to extract for voice answers." Underused in crypto, big AEO impact when added.

FAQPage as AEO weapon: AI engines extract FAQ Q&A blocks for direct answers. Pages with 5+ FAQs and proper FAQPage schema get cited 3x more than pages without. Add FAQs aggressively to all key pages.

The citation metadata loop: author Person schema with sameAs to verified profiles signals credentialed source. AI engines weight credentialed sources more. Higher citations means more visibility, which builds more authority, which gets you cited more. Schema is the entry point to the loop.

5 schema strategy mistakes

Beyond syntax errors, common strategic mistakes.

Mistake 1: One schema type for the whole site. Treats every page identically. Different pages need different schema. Build a per-template schema strategy.

Mistake 2: Validating once at launch and never again. Schema breaks silently as templates change. Validate on every release.

Mistake 3: Ignoring schema warnings. "Warnings not errors" means rich result eligibility is reduced. Fix warnings, don't skip them.

Mistake 4: Using Yoast/RankMath defaults blindly. Generic SEO plugins emit Article schema everywhere. Override for protocol/token/exchange pages.

Mistake 5: No Speakable schema. Single highest-impact AEO addition for sites that already have FAQ blocks. 5-minute fix.

Schema tools I actually use

Real stack from TG3 client work.

Schema.org Validator for syntax. Free.

Google Rich Results Test for eligibility. Free.

Bing Markup Validator for stricter parsing. Free, important for ChatGPT.

Crawlux Token Schema Audit module for crypto-specific schema validation that generic tools miss.

JSON-LD Generator (technicalseo.com) for boilerplating schema. Free.

Schema App (paid, $99/mo) for enterprise-scale schema management with versioning.

How Crawlux fits in schema work

Token Schema Audit is the dedicated module.

Crypto-specific validation: validates FinancialProduct, Cryptocurrency, CryptoExchange schema with Web3-aware logic. Generic SEO tools miss issues specific to crypto.

Schema gap analysis: compares your schema against best-practice patterns for your page type. Suggests missing properties.

AEO citation correlation: tracks which schema types correlate with AI engine citation rates on your pages.

Validation across engines: runs Schema.org, Google Rich Results and Bing checks in one report.

Per-page schema audit: per-URL breakdown of schema present, missing or wrong type. Actionable per-page fix list.

Free tier: Token Schema Audit on one domain. Module details.

30-day schema overhaul

Sequenced. Skip steps already done.

Days 1-3: Audit baseline. Run Crawlux Token Schema Audit. Document current schema by page template.

Days 4-10: Schema by page type. Migrate protocol pages to FinancialProduct, token pages to Cryptocurrency, exchange pages to CryptoExchange, blog posts to Article + Person and comparison pages to Review.

Days 11-17: FAQ density push. Add 5+ FAQs to every key page with FAQPage schema. Use AlsoAsked and Reddit for real questions.

Days 18-24: Speakable schema. Add Speakable wrapping direct-answer paragraphs and FAQ blocks. Single highest AEO enable.

Days 25-30: Validate and monitor. Run Schema.org, Bing and Google validators. Fix all errors and warnings. Set up monitoring for schema breakage. Re-run audit and compare to baseline. Schema fixes typically lift traffic 30-50% in 60 days.

From the TG3 client roster

Audit your site against this guide

The Crawlux audit modules below test specific patterns from this guide on your site automatically.

About AB

Compare specific token schema pairs

Detailed head-to-head comparisons for the protocols, projects and tools covered in this guide.

Sources and methodology

This guide synthesizes findings from 200+ Web3 site audits conducted at TG3 Agency since 2017, plus public data verified against the sources below. Last verified May 3, 2026.

• [01]DefiLlama · TVL, volume and protocol metrics
• [02]CoinGecko · Token price, supply and market data
• [03]Schema.org · Structured data specification
• [04]Google Search Central · Structured data implementation guide

This guide is for informational purposes. The crypto SEO landscape changes quickly. Re-run audits quarterly.

Got feedback or a different take?

Drop your perspective below. We read every comment.