The GEO Audit Checklist: 25 Things to Check

An AI visibility problem is almost always a diagnosis problem before it is a tactics problem. This 25-point audit is the exact checklist PRESENSS runs before recommending any campaign work. Score each check pass/fail and prioritize by impact rating.

Time to run: 3–4 hours for a small site, 1–2 days for a large one.

Section 1 — Technical foundation (8 checks)

1. llms.txt is published at /llms.txt and returns 200. — Critical

2. robots.txt allows GPTBot, OAI-SearchBot, PerplexityBot, ClaudeBot, Google-Extended, BingBot. — Critical

3. WAF / Cloudflare bot management is not challenging AI crawlers. — Critical

4. Organization JSON-LD is present on the homepage with full sameAs array. — High

5. FAQPage schema is deployed on every real Q&A block. — High

6. Article schema with author, datePublished and dateModified is present on every editorial page. — High

7. Core Web Vitals: LCP under 2.5s, no CLS regressions on primary landing pages. — Medium

8. HTTPS everywhere; consistent canonical tags; no duplicate content across subdomains. — Medium

Section 2 — Entity consistency (6 checks)

9. Brand name is spelled and capitalized identically across all Tier 1/2 sources. — Critical

10. Founding year is identical across Wikipedia, Crunchbase, LinkedIn and your site. — High

11. Public pricing on your site matches G2, Capterra and any other price-listing profile. — High

12. Wikidata entity exists with sourced claims on core properties. — High

13. LinkedIn company page description matches your canonical brand description verbatim. — Medium

14. Google Business Profile is claimed, completed and category is correct. — Medium

Section 3 — Citation audit (5 checks)

15. You have at least 25 substantive G2 reviews (or Capterra equivalent), and at least 5 in the past 90 days. — High

16. Recent Reddit discussions in your category mention your brand at least occasionally. — High

17. You have at least one earned mention in a Tier 1 media source (Reuters, Bloomberg, TechCrunch, or top vertical trade press) in the past 12 months. — High

18. You are cited by at least three category-relevant YouTube reviewer channels or podcasts. — Medium

19. You have a Wikipedia article (if notability supports it) and the article is factually accurate. — Medium

Section 4 — Content gaps (6 checks)

20. You have a dedicated use-case page for each of your top 3 buyer personas. — High

21. You have a "vs" comparison page against each of your top 3 named competitors. — High

22. You have an FAQ page or FAQ blocks answering your buyers' 10 most-common evaluation questions. — High

23. Pricing is publicly listed on a canonical pricing page with SoftwareApplication or Product schema. — High

24. You have llms-full.txt published for deeper AI ingestion (optional but valuable). — Medium

25. You have run a persona-query PRESENSS scan in the past 30 days. — High

Scoring guide

Count the checks you pass. Interpret as follows:

| Passed | State | Recommended next step |

|---|---|---|

| 22–25 | Mature GEO posture | Maintenance cadence + earned coverage program |

| 17–21 | Solid foundation, tactical gaps | Close remaining High-impact items first |

| 11–16 | Meaningful gaps | Run the 30-Day GEO Sprint |

| 0–10 | Under-instrumented | Start with the 10 Quick Wins |

FAQ

Which check is the single most common failure?

Check 3 — WAF / Cloudflare bot management silently challenging AI crawlers. Countless brands fail this without realizing it. Test each AI user-agent by fetching a real page and inspecting the response code.

How often should we re-run this audit?

Quarterly. Entity properties drift, review recency decays, and content ages faster than most teams update it.