How to Conduct a GEO Audit: Step-by-Step Guide (2026)

1 Define Your Query Set

The foundation of any GEO audit is a well-defined set of queries that your target customers use when researching your category.

How to Build Your Query Set:

1. Start with your money keywords: The terms that drive the most business value
2. Add comparison queries: "[Your Brand] vs. [Competitor]" and "best [product] for [use case]"
3. Include question-based queries: "How to choose [product]", "what is the best [service]"
4. Cover the full funnel: Awareness, consideration, decision
5. Aim for 50-100 queries: Enough for statistical significance, manageable for manual testing

# Core Routes
- best train London to Edinburgh
- London to Leeds train times
- fastest train London to York
- London to Newcastle train

# Comparison Queries
- LNER vs Lumo vs Avanti
- best train operator East Coast
- which train has best first class
- most reliable train UK

# Question-Based
- how to book cheap train tickets
- what is Azuma train
- do LNER trains have WiFi
- best quiet carriage train
                            

2 Measure Current Inclusion Rate

Now you need to test each query across AI platforms and calculate your baseline Inclusion Rate.

Manual Method:

1. Test each query on ChatGPT, Gemini, Claude, Copilot, and Perplexity
2. Record whether your brand appears in the response
3. Calculate Inclusion Rate = (Appearances ÷ Total Queries) × 100
4. Track sentiment (positive/neutral/negative) when you appear
5. Note whether you're linked or just mentioned

| Query | ChatGPT | Gemini | Claude | Copilot | Perplexity | Avg |
|-------|---------|--------|--------|---------|------------|-----|
| best train London-Edinburgh | ✓ | ✓ | ✗ | ✓ | ✓ | 80% |
| LNER vs Lumo | ✓ | ✓ | ✓ | ✓ | ✗ | 80% |
| London-Leeds train times | ✗ | ✓ | ✗ | ✓ | ✗ | 40% |
| ... | | | | | | |

Overall Inclusion Rate: 23%
                            

3 Audit Schema Completeness

AI platforms rely heavily on structured data to understand your offerings. Without complete schema, even great content may be ignored.

Schema Checklist:

Tools for Schema Validation:

• Google Rich Results Test: Check if your schema is valid
• Schema.org Validator: Official validation tool
• Ahrefs/Semrush: Site-wide schema audits

4 Analyze Information Gain

Information Gain — introduced in the Princeton GEO research — measures how unique and valuable your content is. Content that merely repeats common knowledge has near-zero citation probability.

Information Gain Scoring Framework:

How to Improve Information Gain:

• Conduct original surveys and publish the results
• Share internal data (with privacy considerations)
• Interview industry experts and quote them directly
• Develop proprietary frameworks or methodologies
• Publish case studies with specific metrics

5 Assess Comparison Content

Comparison queries like "[Brand] vs. [Competitor]" are where purchase decisions happen. If you don't have dedicated comparison content, AI will cite competitors who do.

Comparison Content Checklist:

# LNER vs. Lumo vs. Avanti: London to Edinburgh Comparison

| Feature | LNER | Lumo | Avanti |
|---------|------|------|--------|
| Journey Time | 4h 20m | 4h 30m | 4h 45m |
| Price (Advance) | from £42 | from £30 | from £55 |
| WiFi | Free | Free | Free |
| Food Service | Full restaurant | Café bar | First Class only |
| First Class | Yes | No | Yes |
| Quiet Carriage | Yes | Limited | Yes |
| Trustpilot | 4.2/5 (15k) | 4.0/5 (2k) | 3.9/5 (8k) |

[Book LNER] · [LNER Azuma guide] · [LNER first class review]
                            

6 Platform-Specific Deep Dive

Different AI platforms have different preferences. A platform-specific analysis reveals where you're winning and losing.

7 Competitor Benchmarking

You need to know not just where you stand, but how you compare to competitors.

Benchmarking Process:

1. Identify 3-5 key competitors
2. Run the same query set for each competitor
3. Calculate their Inclusion Rate by platform
4. Analyze their schema and content
5. Identify gaps where they appear and you don't

| Competitor | Overall IR | ChatGPT | Gemini | Key Gaps |
|------------|------------|---------|--------|----------|
| LNER | 23% | 28% | 21% | Comparison content |
| Lumo | 31% | 29% | 34% | Schema incomplete |
| Avanti | 27% | 25% | 29% | Review authority |

Your opportunity: Create LNER vs Lumo comparison page
                            

8 Prioritize Opportunities

Now you have a list of gaps. Not all are equally important. Prioritize based on:

• Impact: How much will fixing this increase Inclusion Rate?
• Effort: How long will it take to implement?
• Query volume: How many customers search this way?
• Intent: Is this a high-purchase-intent query?

| Opportunity | Impact | Effort | Priority |
|-------------|--------|--------|----------|
| Add schema to product pages | High | Low | 🟢 High |
| Create LNER vs Lumo comparison | High | Medium | 🟢 High |
| Update 3 blog posts | Low | Low | 🟡 Medium |
| Rewrite 20 product descriptions | Medium | High | 🔴 Low |

Next: Schema implementation (2 weeks)
                            

9 Implement & Monitor

Execute your roadmap and track results.

Implementation Tips:

• Start with high-impact, low-effort items first
• Track Inclusion Rate weekly after changes
• Document what works for future audits
• Set up alerts for significant changes

10 Repeat Quarterly

AI platforms update frequently. Competitors improve. Your content ages. A quarterly audit ensures you stay ahead.

Quarterly Audit Checklist:

• Re-run all 50-100 queries
• Check for new competitors
• Review platform updates (new features, algorithm changes)
• Update your roadmap
• Celebrate wins!