Has anyone done a GEO audit? Trying to figure out where to start with AI search visibility

I do these professionally. Here’s the framework I use.

The 7-Step GEO Audit Process:

1. Prompt Discovery Map questions your target audience asks AI. Not keywords - actual questions. Think buyer journey:

• Awareness: “What is [problem]?”
• Consideration: “Best solutions for [problem]”
• Decision: “[Your brand] vs [competitor]”

2. Cross-Platform Visibility Assessment Test your prompts across:

• ChatGPT (with and without browsing)
• Perplexity
• Claude
• Google AI Overview
• Gemini

Document: Are you mentioned? What position? How are you described?

3. Content Structure Analysis Evaluate how AI-extractable your content is:

• Clear headings matching common questions?
• Concise, citable statements?
• Question-answer formatting?
• Semantic HTML structure?

4. E-E-A-T Signal Audit Check expertise indicators:

• Author bios with credentials
• Expert quotes and citations
• Trust signals (about page, contact info)
• Evidence of real-world experience

5. Structured Data Validation Review schema markup:

• Organization schema complete?
• Article schema with author details?
• FAQ schema on relevant pages?
• Valid JSON-LD without errors?

6. Technical Accessibility Verify AI crawlers can access your content:

• Robots.txt allows GPTBot, ClaudeBot, PerplexityBot
• No aggressive rate limiting
• Server-side rendering for key content
• Fast TTFB (<200ms ideal)

7. Competitor Benchmark Compare your visibility to competitors:

• Who gets cited for your target prompts?
• What content structure do they use?
• What authority signals do they have?

Deliverable: GEO scorecard with scores per component, priority actions, and baseline metrics for tracking improvement.

This is exactly what I needed. Few follow-up questions:

1. How long does a thorough audit take?
2. What tools do you use for the visibility assessment?
3. What does the scorecard look like?

Adding technical depth to James’s framework.

Technical Accessibility Deep Dive:

Robots.txt check: Many sites accidentally block AI bots. Check for:

User-agent: GPTBot
Disallow: /

This kills your ChatGPT visibility.

JavaScript rendering: AI crawlers mostly can’t execute JavaScript. If your content loads via JS:

• It’s invisible to most AI systems
• Implement server-side rendering
• Or pre-render critical pages

Server response times: AI crawlers are impatient. Check:

• TTFB < 200ms ideal
• Total page load < 2 seconds
• No aggressive rate limiting on bots

Canonicalization: Ensure AI systems see correct canonical versions. Duplicate content confuses citations.

Structured data validation:

Use Schema.org validator, not just Google’s Rich Results Test. AI systems use schema.org broadly, not just Google’s subset.

Common issues:

• Invalid JSON-LD syntax
• Missing required properties
• Inconsistent Organization schema across pages
• No author details in Article schema

The quick technical check:

View your page source. Search for “application/ld+json”. Is it there? Is it comprehensive? That’s your starting point.

Content structure perspective for the audit.

What AI-extractable content looks like:

Good:

## How does [X] work?

[X] works by [clear 1-2 sentence explanation]. The process involves [specific steps or components].

Bad:

## Overview

In today's complex landscape, understanding [X] is more important than ever. Let's explore the various aspects...

AI needs citable statements, not throat-clearing.

Audit checklist for content:

• H2s match questions users ask AI
• First paragraph answers the question directly
• Key facts in standalone sentences (not buried in paragraphs)
• Tables summarizing comparisons or processes
• Lists of specific steps, features, or criteria
• Clear definitions near relevant terms
• Credible sources cited within content

The extraction test:

Ask ChatGPT your target questions. Look at what it cites.

Then ask: Is our content structured to be cited that way? If competitors’ content is more extractable, that’s why they’re winning.

Adding the E-E-A-T audit component.

What to evaluate:

Experience:

• Do you show real-world use of products/services?
• Customer stories and case studies?
• First-person expertise in content?

Expertise:

• Author credentials clearly displayed?
• Expert contributors credited?
• Depth of knowledge demonstrated?

Authoritativeness:

• Brand mentioned on authoritative sites?
• Industry recognition and awards?
• Media coverage and PR presence?

Trustworthiness:

• Clear contact information?
• Privacy policy, terms of service?
• Customer reviews and testimonials?
• Secure site (HTTPS)?

The assessment matrix:

For each E-E-A-T component, score:

1. Does this signal exist?
2. Is it prominent/findable?
3. Is it consistent across the site?
4. Is it verifiable/credible?

The weak areas become your optimization priorities.

Measurement framework for ongoing GEO tracking.

Baseline metrics to establish:

Visibility metrics:

• Citation frequency (mentions per 100 prompts)
• Position distribution (1st mention vs 5th)
• Platform coverage (which AI systems cite you)
• Sentiment (how you’re described)

Competitor benchmarks:

• Share of voice vs top 3 competitors
• Citation contexts where they win

Content performance:

• Which pages get cited most
• Content types that perform (blog vs product vs FAQ)

How to track over time:

Weekly:

• Manual spot checks on key prompts
• Am I Cited dashboard review

Monthly:

• Full visibility score trending
• Competitor share of voice changes
• Content performance analysis

Quarterly:

• Comprehensive re-audit
• Strategy adjustment based on data

The reporting template:

Show leadership:

1. Current visibility score (and trend)
2. Competitor comparison
3. Actions taken
4. Next priorities

GEO becomes measurable and accountable.

Client delivery perspective.

How we present GEO audits:

Executive Summary (1 page):

• Current visibility score
• Key gaps vs competitors
• Top 3 priority actions
• Expected timeline for improvement

Detailed Findings (10-15 pages):

• Each audit component with scores
• Specific evidence and examples
• Competitor comparison data
• Prioritized recommendations

Technical Appendix:

• Schema validation results
• Crawler access audit
• Page-by-page analysis for key content

What gets leadership attention:

Visual comparisons. “When users ask ChatGPT about [topic], Competitor A is mentioned first 70% of the time. We appear 15% of the time, usually 3rd or later.”

That gap is immediately understood and actionable.

Budget-conscious approach for smaller teams.

The DIY GEO Audit:

Phase 1: Quick Visibility Check (2 hours)

• List 20 key prompts your audience asks
• Test each on ChatGPT and Perplexity
• Document: mentioned? position? context?
• Do same for top 2 competitors

Phase 2: Content Review (4 hours)

• Pick 10 most important pages
• Check each for question-answer structure
• Verify author credentials visible
• Note gaps in extractable content

Phase 3: Technical Basics (2 hours)

• Check robots.txt for AI bot access
• Validate schema with Google’s tool
• Check page speed basics

Free tools:

• ChatGPT and Perplexity (free tiers for testing)
• Google Rich Results Test (free)
• PageSpeed Insights (free)
• Schema.org validator (free)

When to upgrade:

If manual testing reveals significant gaps, invest in proper monitoring tools like Am I Cited. The automation saves time and catches changes you’d miss manually.

Start scrappy, scale as needed.

How GEO audit fits with traditional SEO audit.

The integration approach:

Don’t run separate audits. Extend your SEO audit to include GEO:

Traditional SEO audit adds:

Technical section:

• Add AI crawler access check
• Add schema comprehensiveness review

Content section:

• Add extractability assessment
• Add question-answer format check

Authority section:

• Map E-E-A-T signals explicitly
• Add brand mention analysis

New section: AI Visibility

• Cross-platform citation check
• Competitor visibility benchmark
• Platform-specific gaps

Why integration works:

Many fixes help both. Good schema helps Google AND AI systems. Fast sites rank AND get crawled by AI. Quality content ranks AND gets cited.

The marginal effort to add GEO to an SEO audit is maybe 20%. The insight value is 100% more.

This thread is gold. Thank you all.

My GEO audit plan:

Week 1:

• DIY visibility check (Tom’s approach)
• Identify top 3 gaps

Week 2:

• Technical audit (schema, crawler access)
• Content structure review for key pages

Week 3:

• Set up Am I Cited for ongoing monitoring
• Competitor benchmark

Week 4:

• Compile scorecard (James’s framework)
• Present to leadership with priorities

Deliverable for leadership:

• Executive summary with visibility score
• Competitor comparison (visual)
• Prioritized action plan
• Proposed monitoring approach

Success metric: Establish baseline visibility score, then show improvement over next quarter.

Really appreciate the frameworks and practical advice!