Most brands have no idea how they appear — or fail to appear — in AI-generated answers. They have not tested what ChatGPT says about them, they do not know whether Perplexity is citing their content, and they have not assessed whether their brand is a coherent entity in AI knowledge systems. This audit changes that.
The 40-point GEO audit is organized into four categories that mirror the signal hierarchy AI systems use to evaluate brands: entity signals, technical signals, content signals, and off-site authority signals. Each check is worth one point. Score yourself honestly and use the scoring tiers to prioritize your roadmap.
How to use this audit: Work through all 40 checks and mark each one you can honestly confirm is in place. Do not give yourself credit for partial implementation. Your final score will tell you which tier you are in and where to focus first. If you want a scored audit conducted by our team with specific recommendations, book a free call at the end of this page.
Typical GEO Audit Score Distribution — How brands score on first audit (n=200+ audits)
77% of brands score below 25 on their first GEO audit — suggesting significant untapped AI visibility opportunity for most organizations.
Category 1: Brand Entity Signals (10 points)
Category 01
Brand Entity Signals
-
Wikidata item exists and is populatedYour organization has a Wikidata entry (Q-number) with at minimum: instance of organization, official website, inception date, country, and industry.
-
Wikipedia article exists or notability criteria are clearly metYour brand has a Wikipedia article, or you have documented coverage in multiple independent, reliable sources sufficient to qualify for one.
-
Google Knowledge Panel is live and claimedYour brand has a Knowledge Panel in Google Search, and it has been claimed and verified via Google Search Console.
-
Organization schema on homepage with full sameAs arrayYour homepage includes JSON-LD Organization schema with sameAs pointing to Wikidata, LinkedIn, Crunchbase, and other authority profiles.
-
Brand name is identical across all major directoriesLinkedIn, Crunchbase, Google Business Profile, Glassdoor, and your website all use the exact same brand name format — no variations.
-
Website URL is consistent across all external listingsAll directory and profile listings use the same canonical URL (same protocol, www/non-www, no trailing slash inconsistencies).
-
Industry and category classification is consistentYour primary industry, category, and business type are described the same way across LinkedIn, Crunchbase, Google Business Profile, and G2/Capterra.
-
Crunchbase or equivalent authority company profile existsYour company has a complete Crunchbase profile (or equivalent: CB Insights, PitchBook) with funding, leadership, and category data.
-
LinkedIn Company Page is complete and actively maintainedCompany Page has a filled-in description, correct industry, website URL, founded year, company size, and has posted content in the last 30 days.
-
AI test: brand is described accurately in ChatGPT and PerplexityWhen you ask "tell me about [brand name]" in ChatGPT and Perplexity, the response is accurate, complete, and mentions the right category and value proposition.
Category 2: Technical & Schema Signals (10 points)
Category 02
Technical & Schema Signals
-
robots.txt allows all major AI crawlersPerplexityBot, GPTBot (OpenAI), Google-Extended, Anthropic-AI, and Meta-ExternalAgent are all permitted — or you have a deliberate policy to allow/block specific ones.
-
Article or BlogPosting schema on all content/blog pagesEvery blog post and article page has Article or BlogPosting JSON-LD schema with datePublished, dateModified, author, and headline.
-
FAQPage schema deployed on core product and topic pagesYour main product pages, service pages, and key blog posts include FAQPage JSON-LD with 3–5 question-answer pairs matching real user queries.
-
BreadcrumbList schema implemented sitewideAll pages with a clear hierarchy (not just the homepage) include BreadcrumbList JSON-LD, helping AI models understand your site structure.
-
Product or SoftwareApplication schema on product pagesYour product/software pages use the appropriate schema type with name, description, offers, and aggregateRating where applicable.
-
XML sitemap is current and includes lastmod datesSitemap is submitted to Google and Bing Search Console, includes all key pages, and uses lastmod dates that accurately reflect content updates.
-
Core Web Vitals pass on key landing pagesLCP under 2.5s, INP under 200ms, CLS under 0.1 on your homepage and top 10 pages — slow pages are deprioritized in AI retrieval pipelines that rely on Bing/Google indexing.
-
Canonical tags are correctly implemented across all pagesEvery page has a self-referencing canonical tag (or correctly points to the preferred URL) — no conflicting canonicals that could confuse AI crawlers about which version of a page to index.
-
Internal linking architecture connects topically related contentYour key topic clusters are interconnected via contextual internal links — AI models use link structure to understand topical authority and entity relationships within your site.
-
No significant crawl errors on priority pagesGoogle Search Console shows no 4xx or 5xx errors on your top 20 pages — pages that return errors cannot be indexed or cited.
Category 3: Content Signals (10 points)
Category 03
Content Signals
-
Core pages answer the primary question in the first 200 wordsYour homepage, product pages, and top blog posts lead with the direct answer — no lengthy preambles before the content that actually answers what visitors (and AI) need to know.
-
Key content has been published or meaningfully updated in the last 90 daysAt least 40% of your top 20 pages by traffic have a dateModified within the last 90 days — with actual content changes, not just date-stamp updates.
-
Long-form authority guides exist for each core topic clusterYou have at least one comprehensive (2,000+ word) guide for each major topic your brand wants to be cited on — covering definitions, best practices, and examples.
-
Content cites specific statistics with named sourcesYour content uses precise data points with sourced references (not vague claims like "many companies") — AI models replicate the citation patterns of their training content.
-
Use case and vertical-specific content exists for each target segmentFor each major customer segment or vertical you serve, there is dedicated content explicitly naming that segment and explaining why your product fits their specific needs.
-
Comparison and alternatives content covers key competitive queriesYou have pages that directly address "[Your brand] vs. [Competitor]" and "alternatives to [Competitor]" queries with genuine, specific comparative analysis.
-
Content demonstrates E-E-A-T signals throughoutAuthors are named and credentialed where appropriate, content references first-hand experience or proprietary data, and About/Team pages substantiate organizational expertise.
-
Customer case studies include specific, named, verifiable outcomesYour case studies name the customer (or a named segment), specify the exact metric improvement, and include a direct quote — vague social proof does not generate AI citations.
-
Glossary or definitional content covers key terms in your categoryYou have structured definitional content for the core terms in your industry — this is the content AI models most frequently cite for "what is X" queries.
-
H2 and H3 headings are structured as natural-language questions or direct answersYour content uses question-format or direct-answer headings (e.g., "How does X work?" or "What is X?") that match the phrasing patterns of AI user queries.
Category 4: Off-Site & Authority Signals (10 points)
Category 04
Off-Site & Authority Signals
-
Coverage in 3+ independent, high-authority publications in the last 12 monthsYour brand has been featured (not just mentioned in press releases) in recognized industry publications, tech press, or mainstream media with editorial independence.
-
Your brand is named in at least one analyst report or industry studyGartner, Forrester, IDC, or a credible industry analyst has named your brand in a report, market map, or category overview — these are extremely high-weight training signals.
-
G2 or Capterra profile has 20+ verified reviews with a 4.0+ ratingReview aggregators are heavily crawled and cited by AI. A robust review presence (20+ reviews, 4.0+ stars) significantly increases the probability of appearing in "best [category]" AI responses.
-
Your brand appears in competitor comparison pages on well-known competitor sitesWhen competitors or comparison sites mention alternatives or competitors in your space, your brand appears — indicating recognized category membership in training data.
-
Thought leadership content is published on external platformsExecutives or team members have bylined articles on Medium, Substack, Forbes Councils, LinkedIn articles, or industry publications — extending your brand's authoritative voice beyond your own domain.
-
Podcast appearances or expert quotes in media in the last 12 monthsPodcast transcripts and quoted expert commentary in articles provide contextually rich, natural-language brand mentions that AI models weight as expertise signals.
-
Backlink profile includes topically relevant high-DA sitesAhrefs or Semrush shows 10+ referring domains with DR/DA above 60 that are topically relevant to your category — these feed the retrieval layer of AI search systems.
-
If you publish proprietary research or data, it has been cited by external sourcesOriginal data, surveys, or reports you publish have been cited or referenced by external publications — co-citation with your original research is a powerful authority signal.
-
Your brand appears organically in community and forum discussionsReddit, Hacker News, Quora, and industry-specific forums contain organic mentions or recommendations of your brand — these communities are actively scraped into AI training data.
-
Partner co-mentions link your brand to recognized names in your ecosystemIntegration partners, agency partners, or technology alliances mention your brand on their sites alongside their own — co-citation with trusted brands transfers authority in AI knowledge models.
Your Score: What It Means
Add up your total checks. Each completed item is worth one point, for a maximum score of 40.
0 – 12
Critical Gaps
AI models have minimal signal to identify and cite your brand. You are likely invisible in AI-generated answers for your core queries. Start with Category 1 and Category 2 immediately — entity foundation and technical prerequisites must come first.
13 – 24
Developing
You have the foundations in place but significant gaps remain. AI models may cite you occasionally but inconsistently. Focus on whichever category has the most unchecked items — the marginal checks in your weakest area will yield the highest lift.
25 – 34
Strong
Your AI search visibility is substantially above average. You are appearing in most relevant AI answers. Focus on the specific unchecked items in Category 3 and 4 — content depth and off-site authority are typically the differentiators at this stage.
35 – 40
Category Leader
You are in the top 5% of brands for AI search visibility. Your GEO foundation is strong — focus on monitoring emerging AI platforms (Gemini, Grok, Claude web), tracking citation trends, and defending your positions as competitors catch up.
Pro tip on scoring: The 40 checks are not equal in impact. In practice, Category 1 checks (entity signals) have an outsized effect on AI citation probability — a brand with 8/10 entity signals and 4/10 everything else will outperform a brand with 5/10 across all categories. If you have limited time, prioritize entity completeness above all else.
Want a professional GEO audit with specific action items?
Our team will run this full 40-point audit on your brand, score each item, and build you a prioritized roadmap. We also test your AI citation rates across ChatGPT, Perplexity, and Google AI Overviews. Free for qualified brands.
Book Your Free Audit →