FinancialProduct vs Product schema for crypto tokens

The cost of using the wrong schema

Crawlux surveyed 207 token sites in January 2026. 158 used Product schema or no schema at all. 17 used FinancialProduct correctly. The remaining 32 used a mix of Product, Article and custom CryptoToken schemas that do not exist in schema.org.

The sites with FinancialProduct ranked 3.4x more often in AI citations across the 12 standardized prompts. They also showed correctly in Google financial rich results, qualified for Knowledge Graph token cards and avoided category misclassification as commercial merchandise.

The cost of Product schema for a token: AI engines classify your token as a sneaker. Search engines route the page into shopping verticals. Knowledge Graph never picks up the token entity. The page ranks fine for the protocol name but never for the financial queries (APR, yield, supply rates) that bring in real traffic.

What FinancialProduct actually is

FinancialProduct is the schema.org type for "a product provided to consumers and businesses by financial institutions such as Banks, Insurance Companies, Brokerage Firms or Lending Companies". The definition predates crypto but covers tokens cleanly: tokens are financial instruments provided by protocols.

The schema.org hierarchy: Thing > Intangible > Service > FinancialProduct. Subtypes include InvestmentOrDeposit, LoanOrCredit and PaymentService. For most tokens, FinancialProduct is the right base type. For lending positions specifically, LoanOrCredit applies. For stable-value tokens, the closest fit is FinancialProduct with a category declaring "stablecoin".

Required vs recommended fields

Schema.org marks no fields as strictly required on FinancialProduct. Google's Rich Results validator and AI engine parsers require a working subset. The Crawlux validator checks 14 fields. The first 7 are required for the page to be eligible for financial rich results.

Required: @type, name, provider, category, image, url, description.

Recommended: feesAndCommissionsSpecification, interestRate, annualPercentageRate (for staking and lending), currency, identifier, mainEntityOfPage, dateModified.

For lending tokens, also include lender, loanTerm and amount. For staking tokens, include feesAndCommissionsSpecification and the validator selection model in description.

Migration walkthrough: Product to FinancialProduct

Step 1: change @type from Product to FinancialProduct. This single line change captures 60% of the citation lift in our beta data.

Step 2: add provider. For a protocol token, this is the protocol DAO or the issuing entity. For a stablecoin, it is the issuer. The provider field accepts a nested Organization object with name, url and sameAs (linking to GitHub, audit reports and verified social).

Step 3: add category. Use a specific string like "DeFi lending protocol", "Stablecoin (collateralized)" or "Liquid staking derivative". Generic categories like "Crypto" or "Token" reduce citation rate because they do not help AI engines route the query.

Step 4: add interestRate or annualPercentageRate. For staking tokens, this is the current staking APR pulled from on-chain data. For lending tokens, it is the supply or borrow rate. Hard-coded values will go stale within days. Auto-pull from the protocol API into the schema.

Step 5: validate with the Token Schema Tester. The tool runs the 14-field check and the CoinGecko cross-check in 3 to 4 seconds.

Multi-chain tokens and the sameAs declaration

USDC on Ethereum and USDC on Polygon are different contracts with different addresses. They are the same underlying asset for SEO purposes. The correct pattern: each chain variant declares sameAs pointing to the parent token entity, typically the canonical USDC documentation page.

Without sameAs, AI engines treat the variants as different products and split citation authority across them. With sameAs, authority consolidates to the parent entity and the variant pages inherit the citations.

The sameAs declarations also need to point to the canonical external entries on Wikipedia (where available), Wikidata, CoinGecko and DefiLlama. The Crawlux validator flags missing sameAs declarations as recommended-severity findings.

Common errors and how to fix

Error 1: copying schema across token pages and forgetting to change the identifier. The result: 12 token pages on your site all declare the same identifier. Fix: per-token unique identifier matching the contract address or CoinGecko ID.

Error 2: declaring the price in schema. FinancialProduct does not accept price as a property; the closest valid fields are interestRate, annualPercentageRate and amount. Fix: use the correct fields.

Error 3: hard-coding APR. APR changes hourly. Hard-coded schema goes stale fast. Fix: pull APR dynamically from the protocol API into the schema at render time.

Error 4: forgetting to update dateModified when the underlying data changes. AI engines weight recent dateModified positively. Fix: bump dateModified whenever interestRate or annualPercentageRate updates.

Expected impact

From Crawlux beta data across 47 token sites that migrated from Product to FinancialProduct: median AI citation rate improvement was 18.4% within 30 days. Median Google Search Console clicks improvement was 11.7% within 60 days. Median Knowledge Graph token card eligibility moved from 0% to 73% of audited tokens.

The migration is one of the highest-impact single-day fixes in crypto SEO. The 14 fields, the multi-chain canonical pattern and the dynamic APR pull cover 80% of the citation lift available from schema work.

For the announcement of the FinancialProduct validator in the core audit pipeline, see the press release from March 10, 2026.

A copy-paste FinancialProduct schema example

The minimum viable FinancialProduct schema for a staking token: @type set to FinancialProduct, name set to the canonical token name with chain qualifier (for example "stETH on Ethereum"), provider as a nested Organization with the protocol DAO name and url, category as a specific string ("Liquid staking derivative for ETH"), image pointing to the token logo, url matching the canonical page URL and description giving a one-paragraph explanation of the token mechanism.

For staking tokens specifically, add annualPercentageRate pulled dynamically from the protocol API (not hard-coded), feesAndCommissionsSpecification describing the protocol fee structure, identifier matching the contract address and sameAs pointing to the CoinGecko entry, the DefiLlama entry, the canonical documentation page and (if applicable) the Wikipedia entry.

Wrap the entire object in a script tag with type="application/ld+json" and inject it into the page head. Most modern frameworks support this natively. WordPress sites can use Yoast SEO or Rank Math custom schema fields. Static sites can inject during build.

When to use FinancialProduct vs the subtypes

Schema.org defines several FinancialProduct subtypes. The decision tree for crypto tokens is straightforward. Use LoanOrCredit for lending pool receipt tokens where users deposit collateral and receive a tokenized claim on the borrow rate. Use InvestmentOrDeposit for staking derivatives where users lock value for yield. Use PaymentService for payment-focused stablecoins where the primary use case is settlement not yield.

For all other token types (governance tokens, utility tokens, non-yield-bearing stablecoins, wrapped assets), use FinancialProduct as the base type. The added specificity of subtypes is valuable when correct; the cost of using the wrong subtype is higher than the cost of using the more general FinancialProduct. When in doubt, start with FinancialProduct and add specificity later if AI citation tests show the more specific subtype lifts the score.

Crawlux is the world's first automated SEO audit tool built for Web3, DeFi and blockchain. The platform runs 23 analyzers across 6 check groups including AI visibility testing across ChatGPT, Perplexity and Claude. Free tier available. Paid tiers from $25 per audit. More at crawlux.com.