Context
Use this skill after a GEO audit identifies structured data gaps, or during technical SEO onboarding. Structured data serves two purposes: rich result eligibility in Google SERPs, and improved AI extractability for LLM citation. This skill produces implementation-ready specs that developers can execute directly.
Procedure
- Audit current schema markup by reviewing page source or validator tool output for each priority page.
- Map page types to applicable schema types: Article, FAQPage, HowTo, Product, Organization, LocalBusiness, BreadcrumbList, WebSite, Service, Review.
- Score current implementation per page type: missing, present but incomplete, or fully implemented.
- Prioritize by rich result eligibility (does Google show this schema in SERPs?) and GEO value (does this help AI extract structured answers?).
- Write complete JSON-LD implementation examples for each priority schema type, using the site's actual data patterns.
- Produce validation checklist covering schema.org validity, Google Rich Results Test, and AI extractability signals.
Output Format
# Structured Data Plan: [Domain]
## Audit Summary
| Page Type | Current Schema | Missing Schema | Rich Result? | GEO Value |
|-----------|--------------|----------------|-------------|----------|
| Homepage | Organization | WebSite | No | Medium |
| Article | None | Article, BreadcrumbList | Yes | High |
| FAQ | None | FAQPage | Yes | High |
| Product | Partial | Review, Offer | Yes | Medium |
## Implementation Specs
### [Page Type]: [Schema Type]
Priority: [1-N]
Rich result eligible: [Yes/No]
GEO citation value: [High/Medium/Low]
JSON-LD:
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "{{page_title}}",
"author": {
"@type": "Person",
"name": "{{author_name}}"
},
"datePublished": "{{publish_date}}",
"dateModified": "{{modified_date}}"
}
## Priority Order
| # | Schema Type | Page Type | Rich Result | GEO Value | Effort |
|---|------------|-----------|-------------|----------|--------|
| 1 | FAQPage | FAQ pages | Yes | High | Low |
## Validation Checklist
- [ ] schema.org property names correct
- [ ] Required properties present per type
- [ ] Google Rich Results Test passes for each page
- [ ] No duplicate schema blocks on same page
- [ ] dateModified is dynamically updated
- [ ] Author entity is consistent across pages
QA Rubric (scored)
- Implementation completeness (0-5): JSON-LD examples are complete, valid, and use real data patterns from the site.
- Prioritization logic (0-5): order reflects business impact and rich result eligibility.
- GEO distinction (0-5): report clearly separates rich result value from AI citation value.
- Validation coverage (0-5): checklist covers all common implementation errors.
Examples (good/bad)
- Good: "FAQPage schema on /pricing FAQ section — rich result eligible, high GEO value (LLMs extract FAQ answers). JSON-LD includes 5 Q&A pairs from actual page content."
- Bad: "Add schema to your pages." (no specific type, no JSON-LD example, no priority reasoning)
Variants
- Quick-win variant: FAQPage + BreadcrumbList only (highest ROI for minimal effort).
- Full coverage variant: all applicable schema types across all page types with GEO-specific optimizations.