GEO Audit · Generative Engine Optimization
GEO Audit: What It Is, What It Covers, and How to Run One
A GEO audit tells you why your brand is not being recommended in AI-generated answers and what to fix first. It covers entity clarity, crawler access, content extractability, structured data, citation sources, and per-engine mention rates. This page walks through each area, gives you a free checklist, and shows you the process from start to fix list.
What is a GEO audit
A GEO audit measures your recommendation presence in AI answers
Generative engine optimization is the work of getting your brand named in the synthesized, multi-brand recommendation answers that AI tools give when a buyer asks "what's the best [category] platform for [use case]?" Those answers name three to five brands. A GEO audit measures your presence in them: how often you're named, on which engines, for which prompts, and why competitors appear where you don't.
The GEO audit is the first step of any GEO program because you can't fix what you haven't measured. It tells you exactly where you're invisible, which audit area is causing the gap, and what to fix in what order. Without the audit, you're guessing.
A GEO audit is different from an AEO audit. A GEO audit focuses on synthesized multi-brand recommendation presence. An AEO audit focuses on direct-answer extraction for single-brand factual queries. Both matter, but they diagnose different surfaces and require different fixes. See our AEO audit guide if direct-answer surfaces are your primary concern.
What a GEO audit covers
Six audit areas that determine your AI recommendation share
A complete GEO audit covers six distinct areas. Each one can independently block your brand from appearing in AI answers, so the audit has to check all six before you can prioritize fixes.
Entity clarity audit
AI engines build a model of your brand from what the web says about you. We check that your brand name, product category, and core value proposition are described consistently across your own site, LinkedIn, Wikidata, major review platforms, and the third-party sources AI is most likely to read. Inconsistent descriptions cause AI to hedge or skip your brand entirely.
AI crawler access audit
We check your robots.txt for blocks on GPTBot, ClaudeBot, PerplexityBot, and Google-Extended. We verify your llms.txt and llms-full.txt files are present, accurate, and structured so AI indexers can retrieve your most important pages. Blocked crawlers mean even a perfectly optimized site is invisible to the engines you care about.
Content extractability review
AI models read the first quarter to third of a page and pull self-contained chunks. We review your highest-intent pages for answer placement (is the key fact in the first 200 words?), heading structure (do you use question-based H2s?), and content boundaries (are your answers self-contained or spread across three paragraphs?). We flag every page where extractability is low.
Structured data and schema audit
Schema markup signals to AI what your content is and what it means. We audit your Organization, Service, Product, FAQPage, and BreadcrumbList schema across your key pages, check for validation errors, and identify gaps where schema would improve extraction confidence.
Citation source mapping
Omniscient Digital analyzed 23,387 citations across 240 branded AI prompts and found that only 23% of what AI cites about a brand comes from that brand's own site. We map your current third-party citation sources, identify gaps in review platforms, practitioner communities, analyst coverage, and publication mentions, and prioritize where new citations would move your recommendation share fastest.
Per-engine prompt testing
We run your top 10 to 20 buyer-intent prompts across ChatGPT, Perplexity, Google AI Mode, Gemini, and Claude, recording your mention rate on each. AI engines agree with each other only about 9% of the time, so we report per engine, never a blended score. You see exactly where you're named, where competitors take your spot, and where no brand dominates yet.
How to run a GEO audit
The three-step process
Running a GEO audit yourself takes two to four days for a focused B2B tech site. A professional audit typically takes five to seven business days and produces a more complete citation source map.
Define your prompt set
Write 15 to 30 buyer-intent prompts that represent the questions your buyers ask when researching your category. Include discovery questions, comparison questions, and use-case-specific questions. Exclude branded prompts from the benchmark so the results show organic recommendation share.
Run the audit framework
Work through the six audit areas above: entity clarity, crawler access, extractability, schema, citation sources, and per-engine testing. Document every finding with a severity rating. Flag gaps between what AI says about you and what you want it to say.
Score and prioritize fixes
Group findings into three buckets: technical (fix within days), on-page (fix within weeks), and off-page authority (build over months). Start with the technical fixes because they unlock the rest. Then restructure your highest-intent pages. Then go after the citation sources AI trusts most in your category.
Free GEO audit checklist
14-point GEO audit checklist
Work through each item and flag anything that's missing or inconsistent. Every unchecked item is a potential gap in your AI recommendation presence.
Want the audit done for you?
Our AI visibility audit runs 50 to 100 buyer-intent prompts, maps your citation sources, and delivers a prioritized fix list within five to seven business days.
See the AI Visibility Audit serviceFree vs paid GEO audit
When the checklist is enough, and when it isn't
The checklist works if...
You're early-stage and want to triage before investing in a full audit
You have a technical team that can run the prompt tests and pull the citation data
You want to self-assess before a strategy call so you can ask better questions
Your primary goal is to fix the technical and on-page layers, which the checklist covers well
A professional audit is worth it if...
You need a complete citation source map, not just a list of gaps
You want per-engine mention rates across 50 to 100 prompts, not just 10
You're entering a new category or repositioning and need a clean baseline
You want a prioritized fix list with effort and expected impact estimated for each item
Technical GEO audit
What the technical GEO audit covers
The technical GEO audit is the fastest layer to fix and the most commonly missed. It covers everything that determines whether AI crawlers can reach and parse your site, and whether the structured signals you send them are accurate.
Robots.txt review for GPTBot, ClaudeBot, PerplexityBot, and Google-Extended blocks. Common: marketing teams add broad bot blocks during a spam campaign and never remove them.
Presence, accuracy, and structure. The files should link directly to your most important product, service, and pricing pages, not just your homepage and blog index.
Organization, Service, Product, FAQPage, BreadcrumbList, and Person (for named authors). Validated against Google's Rich Results Test and Schema.org spec.
AI crawlers have short timeouts. Pages that take more than three to four seconds to respond are frequently abandoned mid-crawl. We test on representative bot user agent strings.
Your most important pages need to be reachable in two to three clicks from your homepage. Pages buried deep in your site architecture or only accessible via JavaScript-rendered navigation are often missed entirely.
Related
AEO Audit
What an AEO audit covers: direct-answer extraction and featured-snippet surfaces.
GEO Audit vs SEO Audit
What each type of audit catches, what each misses, and when you need both.
AI Visibility Audit (service)
The professional audit service: 50 to 100 prompts, per-engine scoring, and a prioritized fix list delivered in five to seven days.
GEO audit questions, answered.
Ready for the professional audit?
See where your brand stands in AI answers today.
Our AI Visibility Audit runs 50 to 100 buyer-intent prompts across ChatGPT, Perplexity, Gemini, Claude, and Google AI Mode and delivers a zero-visibility inventory, per-engine breakdown, competitor citation map, and prioritized fix list within five to seven business days.