GEO audit

Summary#

The Free GEO Audit scores how ready a page is to be quoted inside a generative AI answer. It fetches the page, its robots.txt and its llms.txt, evaluates six weighted signals, and returns a 0-100 score with a ranked fix list. It runs at /free-tools/geo-audit with no account.

Purpose#

Generative Engine Optimization is the work of getting your page cited inside an AI answer rather than ranked in a list of links. The mechanics differ from classic SEO in one important way: engines quote passages, not pages. That rewards content shaped as clear, self-contained answers under clean headings, and it punishes a wall of undifferentiated prose no matter how well it ranks on Google.

This audit checks the structural half of that work, which is the half you can measure from HTML.

Overview#

Three fetches run in parallel: the URL you entered, /robots.txt and /llms.txt. Six checks are then scored, each with a fixed weight totalling 100, and the percentage of weight you earned becomes the score.

The check set is deliberately different from AI visibility checker. That tool spreads eleven checks across everything an engine needs to read a page. This one concentrates the weight on the four things that decide whether a passage gets lifted: crawler access, structured data, question-shaped content and heading structure.

Benefits#

• Tells you in one run whether an engine can reach the page at all, which is the precondition for every other GEO tactic.
• Weights answer-shaped structure heavily, so the fixes it surfaces are the ones that change citability rather than ranking.
• Reports the specific finding behind each check, including which crawlers are blocked and how many H2 sections exist.
• Works on any public URL with no account and no credits.

Use Cases#

• Editing a page for AI answers. Run it before and after a rewrite to confirm the restructure actually landed.
• Choosing which page to fix. Run it on your three most commercially important pages and start with the lowest score.
• Explaining GEO to a stakeholder. The six named signals with point values are an easier conversation than an abstract argument about AI search.
• Checking a competitor's approach. See which of the six they have solved.

Requirements#

A public URL returning HTML. Private, loopback and internal hosts are refused, and a page behind a login cannot be read.

Permissions#

Anyone. No account, plan or role. A shared hourly fair-use limit applies per IP address across all free tools.

Cost#

Free. Zero credits.

Navigation Path#

metricvaultai.com → Free tools → GEO Audit

The page is at /free-tools/geo-audit.

Inputs#

Helper line: No signup · No credit card · Tests the live page, robots.txt and llms.txt.

Step-by-Step Guide#

1. Open /free-tools/geo-audit.
2. Enter the page in the field marked example.com. Audit the article you want quoted, not the homepage, unless the homepage is the page in question.
3. Select Run GEO audit. The button reads Checking… and the result area shows Reading the page, robots.txt and llms.txt, and scoring AI readiness….
4. Read the score, then work down Priority fixes from the top. It is sorted by weight.

Reading the Results#

The score ring. 0-100 with a tier badge:

The heading beside the ring reads <N> of 6 AI-readiness checks passed, and the How to read it strip shows the bands 80-100 strong, AI can cite you, 55-79 partly ready and 0-54 at risk.

GEO signals we score. The six checks with their weights:

The last two are compound: both halves must be true to earn the points, and the detail line tells you which half failed. Title and meta reports the title length when the title is the problem and Meta description missing or short (aim 50+ chars). when it is not.

Priority fixes. Failed checks, numbered and sorted by weight, with the same detail text.

The crawler check tests GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot, Google-Extended, CCBot and Applebot-Extended, plus a blanket User-agent: * with Disallow: /.

Examples#

Screenshots#

Tips#

• Question/FAQ format is more forgiving here than in the AI Visibility Checker. A heading written as a question passes it, so this is often the cheapest 16 points on the page.
• Hover the i marker beside a check name for a short explanation of the signal and what a good result looks like.
• The AI crawler access result is origin-wide. Fixing robots.txt lifts every page on the domain at once.
• Re-run immediately after a deploy. Nothing is cached, so the score reflects what is live now.

Best Practices#

• Solve crawler access and structured data first. They are 42 of the 100 points.
• Write the answer in the first paragraph under each question heading, then expand. Engines lift the opening sentences of a section.
• Keep the title inside 15 to 65 characters and the meta description above 50. That is one edit for 14 points.
• Use this alongside AI citation audit. Structure gets you quoted; concrete facts and a named author get you attributed.

Common Mistakes#

• Treating GEO as a ranking hack. The score measures citability. It is not a Google ranking factor and we do not present it as one.
• Publishing an empty llms.txt to win the points. The check also requires 300 words of real content, and a file pointing at nothing helps nobody.
• Auditing the homepage and stopping. Assistants cite specific pages. Audit the pages you want quoted.
• Expecting the page to be saved. Anonymous runs are not stored anywhere.

Limitations#

• One URL per run. No crawl.
• Structural readiness only. No model is queried, so this cannot tell you whether any engine cited you.
• Server-rendered HTML only. Content added by JavaScript after load is invisible to the fetch.
• llms.txt is a proposed convention. Publishing one is cheap and low risk, but no major engine has confirmed that it reads one, and the tool's own on-page guidance says so.
• No export, history, scheduling or recommendations panel. See Free tools, no account needed.

Troubleshooting#

FAQs#

What is a GEO audit? A check of whether your page is structured so a generative engine can lift a clean, attributable answer out of it. This one scores six signals and returns a 0-100 number with fixes.

How is GEO different from SEO? SEO optimizes for a position in a list of links. GEO optimizes for being quoted inside an answer. They overlap heavily, because clean structure and clear content help both, but the target output is different.

Does fixing these signals help my Google ranking too? Some of them, indirectly. A correct title, a single H1, real headings and valid structured data are good classic SEO. We will not promise a ranking change from this score, because it is not a ranking metric.

Which engines does this target? The signals it checks, crawler access, structured data, answer-shaped content and headings, are the ones the major answer engines rely on. It does not query any of them.

How often should I run it? After any structural change to a page or template, and after any robots.txt change. There is no cost, so re-running is free.

What does the paid product add? Repeat measurement of whether the engines actually mention and cite you, competitor comparison, and trend over time. See Visibility Overview and What you get on Free. Inside the app, GEO Content Scorer scores draft content on the same idea before you publish it.