AEO Readiness Audit
Audit your website, APIs, structured data, llms.txt, and execution-layer readiness for AI agents, answer engines, and agentic commerce.
Updated March 9, 2026
A systematic evaluation of your digital infrastructure across two layers:
Read Layer — Can AI systems discover and understand your content?
Execution Layer — Can autonomous agents take action through your infrastructure?
Most organizations score reasonably well on the read layer. Almost none have assessed their execution layer readiness.
Audit scope#
Read Layer Assessment#
Structured Data Coverage — Schema.org markup completeness and accuracy. Entity consistency across pages and platforms. JSON-LD implementation quality.
Content Structure for AI — Semantic HTML quality. Information density versus marketing prose ratio. FAQ and question-answer structures. Content freshness signals.
AI Discoverability — llms.txt presence and quality. robots.txt configuration for AI crawlers. Sitemap structure. Cross-platform consistency.
Authority Signals — E-E-A-T indicators. Source attribution. External validation.
Execution Layer Assessment#
API Surface — Existing API availability and documentation quality. OpenAPI specification presence. Endpoint discoverability for agents. Response format structure.
Protocol Readiness — MCP compatibility assessment. UCP readiness (for commerce). Existing integration patterns. Authentication mechanisms for machine access.
Data Determinism — Structured versus unstructured product and service data. Real-time data availability. Machine-readable constraints. Consistent data across touchpoints.
Action Capability — Available transaction types (buy, reserve, request, subscribe). State transition modeling. Error handling for machine consumers. Verification mechanisms.
Trust Infrastructure — Business identity verification. Data provenance and accuracy. Compliance readiness (GDPR, EU AI Act where applicable). Audit trail capabilities.
Deliverables#
Every audit produces:
AEO Readiness Score — Numeric assessment of read layer and execution layer readiness, benchmarked against your industry.
Gap Analysis — Specific, prioritized list of what is missing, what is broken, and what is working well.
Quick Wins — Changes implementable within days with measurable impact on agent readiness.
Strategic Roadmap — Phased plan for building execution layer capabilities, with timelines calibrated to your industry’s agent adoption curve.
Protocol Compatibility Report — Assessment of which protocols (MCP, UCP, A2A) are relevant for your business and what implementation would require.
Audit formats#
AEO Readiness Audit — Light#
Focused on the read layer with an execution layer overview.
Scope: Website, primary digital properties. Delivery: 5 working days. Best for: Organizations starting to explore AEO, primarily content-driven businesses.
AEO Readiness Audit — Full#
Comprehensive assessment of both layers, including API infrastructure and protocol compatibility analysis.
Scope: Website, APIs, data infrastructure, integration market. Delivery: 10 working days. Best for: E-commerce, SaaS, travel and booking, and any business with existing API infrastructure.
Why now#
The protocols are live. UCP launched in January 2026. MCP is production-ready. Agent-driven commerce is measurable and growing.
Organizations that assess and prepare now build compounding advantages. Those that wait until agent traffic is significant will face more expensive, more disruptive retrofitting.
The Execution Layer — understand what agents need.
AEO vs. SEO vs. GEO — how this fits into your broader strategy.
Evidence rules for a repeatable audit#
Every finding should include the tested URL or endpoint, the request method, the observed output, the expected output, and a severity based on user or business impact. Screenshots are useful for visual defects, but machine facing issues also need raw HTML, response bodies, headers, or schema validation results.
Use the same sample set before and after remediation. A practical sample includes one pillar page, one comparison page, one transactional page, one API action, and one known error path. This avoids a false improvement caused by testing easier pages after the changes.
Severity should reflect consequences. A missing optional description is low priority. A stale price, ambiguous authorization state, or action that can be repeated accidentally is high priority because an agent may make a wrong decision or create a duplicate transaction.
FAQ#
What should an AEO audit test first?#
Start with canonical pages, structured content, important user tasks, and the interfaces that create a state change. Test both successful and failed paths.
Is an AEO audit only a content review?#
No. Content is part of the read layer. A full audit also checks APIs, protocol fit, authorization, deterministic data, errors, confirmations, and auditability.
How often should the audit be repeated?#
Repeat focused tests after material content, schema, API, policy, or workflow changes. Review the broader sample periodically as agent traffic and supported capabilities evolve.