GEO audit template pack: 4 ready-to-use tools
The four tools we pull out on engagements: the 50-point audit grid, the robots.txt that lets AIs through, the schema library to copy, and the 90-day plan. You copy them, you execute.
Optimisé pour être cité dans
This pack brings together the four tools we genuinely use on GEO engagements, delivered here in full and copyable as-is: the 50-point audit grid, a robots.txt that lets AI bots through, a library of ready-to-paste JSON-LD schema, and a 90-day action plan. Not a promise of a PDF — the complete artifacts, right on the page. You copy, you fill in, you fix.
Why templates, not yet another PDF
A guide tells you what to do. A template saves you the hours you'd have spent fumbling to build it yourself. That's the difference, and that's why everything is below, in plain sight, with no download required first.
We didn't create these four tools for this page: we use them on engagements, week after week. The robots.txt is the one we deploy for our clients. The schema library is the one we copy-paste instead of rewriting every time. You're getting tested material, not a mock-up.
The point isn't to replace an engagement. It's to give you enough to move forward on your own, up to the point where an outside eye becomes useful. Work through the four sections in order: diagnosis, technical, markup, execution.
Template 1 — The 50-point GEO audit grid
The GEO audit grid is the exhaustive list of 50 checks we run at the start of an engagement to find out whether a page can be cited by an AI. It's organized into six categories, from technical foundation through to measurement. For each line: what you check, and its impact level. Copy this table into a spreadsheet or a Notion, add a "Status" column (OK / To fix / N/A) and a "Priority" column, then run your key pages one by one. By the end, your gaps show up in black and white, ranked by impact.
A. Technical and indexing
B. Content and intent
C. Passage citability
D. Schema and structured data
E. AI accessibility and off-site
F. Author authority and measurement
Template 2 — The pre-configured AI-friendly robots.txt
The AI-friendly robots.txt is the file that explicitly allows the crawlers of generative engines to read your site. GEO pitfall number one is blocking AIs without realizing it — a restrictive robots.txt, an overzealous security plugin, and your pages become invisible to ChatGPT or Perplexity. Copy the block below to the root of your domain (yoursite.com/robots.txt), replace the domain in the Sitemap and # llms.txt lines, and adapt the Disallow rules to your private directories. It allows legitimate AI bots while keeping control over what you don't want to expose.
# robots.txt — AI-friendly configuration (GEO)
# Place this file at the root: https://yoursite.com/robots.txt
# --- Classic search engines ---
User-agent: Googlebot
Allow: /
User-agent: Bingbot
Allow: /
# --- AI crawlers: training + search ---
# OpenAI / ChatGPT
User-agent: GPTBot
Allow: /
User-agent: OAI-SearchBot
Allow: /
User-agent: ChatGPT-User
Allow: /
# Perplexity
User-agent: PerplexityBot
Allow: /
User-agent: Perplexity-User
Allow: /
# Anthropic / Claude
User-agent: ClaudeBot
Allow: /
User-agent: Claude-Web
Allow: /
User-agent: anthropic-ai
Allow: /
# Google (AI signal / AI Overviews & Gemini)
User-agent: Google-Extended
Allow: /
# Apple, Common Crawl (a frequent LLM source), Amazon, ByteDance
User-agent: Applebot-Extended
Allow: /
User-agent: CCBot
Allow: /
User-agent: Amazonbot
Allow: /
User-agent: Bytespider
Allow: /
# --- Global rules for all other bots ---
User-agent: *
Allow: /
# Block only your genuinely private directories:
Disallow: /admin/
Disallow: /cart/
Disallow: /checkout/
Disallow: /*?*add-to-cart=
# --- Machine references ---
Sitemap: https://yoursite.com/sitemap.xml
# llms.txt: guides AIs to your key content
# https://yoursite.com/llms.txt
Once it's live, test real access: in your server logs, confirm that GPTBot and PerplexityBot actually get a 200, not a 403 returned by your CDN. A permissive robots.txt is useless if a firewall blocks the bot upstream.
Template 3 — The JSON-LD schema library
The JSON-LD schema library is the set of structured-data blocks that Google and the LLMs read first to understand who you are and what a page is about. Each block below is valid and ready to paste into the <head> of your page (inside a <script type="application/ld+json"> tag). Replace every field in square brackets [...] with your real values, and delete the blocks you don't need on a given page. Start with Organization (once, across the whole site) and FAQPage (on your question pages): those are the two with the highest impact.
Organization
To place on every page (or at minimum the homepage). It declares your entity to machines.
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "[YOUR BRAND]",
"url": "https://yoursite.com",
"logo": "https://yoursite.com/logo.png",
"description": "[A SENTENCE DESCRIBING YOUR BUSINESS]",
"email": "[contact@yoursite.com]",
"telephone": "[+1 XXX XXX XXXX]",
"address": {
"@type": "PostalAddress",
"streetAddress": "[STREET]",
"addressLocality": "[CITY]",
"postalCode": "[ZIP CODE]",
"addressCountry": "US"
},
"sameAs": [
"https://www.linkedin.com/company/[YOUR-PAGE]",
"https://www.youtube.com/@[YOUR-CHANNEL]",
"https://x.com/[YOUR-HANDLE]"
]
}
Person (author)
To place on signed pages. It establishes the author's identity and authority — an E-E-A-T signal that AIs cross-reference.
{
"@context": "https://schema.org",
"@type": "Person",
"name": "[YOUR NAME]",
"jobTitle": "[YOUR ROLE]",
"worksFor": {
"@type": "Organization",
"name": "[YOUR BRAND]"
},
"url": "https://yoursite.com/author/[slug]",
"image": "https://yoursite.com/authors/[your-photo].jpg",
"description": "[SHORT BIO WITH CONCRETE EVIDENCE]",
"sameAs": [
"https://www.linkedin.com/in/[YOUR-PROFILE]",
"https://x.com/[YOUR-HANDLE]"
]
}
Article
To place on every article or content page. It tells the machine what it's reading, who wrote it and when.
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "[ARTICLE TITLE]",
"description": "[ONE-SENTENCE SUMMARY]",
"image": "https://yoursite.com/images/[visual].jpg",
"datePublished": "[2026-01-15]",
"dateModified": "[2026-01-15]",
"author": {
"@type": "Person",
"name": "[YOUR NAME]",
"url": "https://yoursite.com/author/[slug]"
},
"publisher": {
"@type": "Organization",
"name": "[YOUR BRAND]",
"logo": {
"@type": "ImageObject",
"url": "https://yoursite.com/logo.png"
}
},
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "https://yoursite.com/[page-url]"
}
}
FAQPage
To place on pages that contain questions and answers. It's the format an AI can reuse directly in an answer — high impact.
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "[YOUR FIRST QUESTION?]",
"acceptedAnswer": {
"@type": "Answer",
"text": "[COMPLETE, SELF-CONTAINED ANSWER, 2 TO 4 SENTENCES.]"
}
},
{
"@type": "Question",
"name": "[YOUR SECOND QUESTION?]",
"acceptedAnswer": {
"@type": "Answer",
"text": "[COMPLETE, SELF-CONTAINED ANSWER, 2 TO 4 SENTENCES.]"
}
}
]
}
BreadcrumbList
To place on deep pages. It helps the machine situate the page within your site hierarchy.
{
"@context": "https://schema.org",
"@type": "BreadcrumbList",
"itemListElement": [
{
"@type": "ListItem",
"position": 1,
"name": "Home",
"item": "https://yoursite.com"
},
{
"@type": "ListItem",
"position": 2,
"name": "[CATEGORY]",
"item": "https://yoursite.com/[category]"
},
{
"@type": "ListItem",
"position": 3,
"name": "[PAGE TITLE]",
"item": "https://yoursite.com/[category]/[page]"
}
]
}
Before publishing, run each block through a structured-data validator (the schema.org Schema Markup Validator or Google's Rich Results Test). Broken schema is useless, and can even hurt you. Also check that every value in the schema matches what's actually visible on the page.
Template 4 — The 90-day action plan
The 90-day action plan is the move from audit to execution, ordered by impact and effort. It splits the fixes into three 30-day phases: first unblock (technical and accessibility), then build (content and markup), finally amplify and measure (authority and steering). Carry each gap from your audit grid into the matching phase, tackle them in order, and don't launch everything at once. Here's the skeleton you copy and adapt to your real gaps.
Days 1 to 30 — Unblock
Put the file from template 2 live, then check in the logs that AI bots actually get a 200.
200 statuses, no stray noindex, presence in the sitemap. This is the bedrock: without it, nothing else gets read.
Confirm the key content is in the raw HTML. If not, prioritize server rendering or pre-rendering.
The two highest-impact markups, to deploy first on the homepage and question pages.
Even a minimal one guides AIs to your key content. Few sites have it — a first-mover advantage.
Days 31 to 60 — Build
On your 5 to 10 business pages, place the answer to the main question within the first 80 words.
Break up the walls of text into self-contained passages, extractable as-is by an LLM.
Align your headings with users' real questions so the machine matches question to answer.
Add Article, Person and BreadcrumbList on the relevant pages, then validate each block.
Named author, bio with evidence, linked Person schema. Authority starts with attribution.
Days 61 to 90 — Amplify and measure
Third-party mentions, Reddit/YouTube presence on your topics, NAP consistency. AIs cross-check before they cite.
Test your 10 to 20 target queries in ChatGPT, Perplexity and Gemini. Note who's cited — you or a competitor.
Isolate visits coming from generative engines. It exists, and it can be measured.
Understand why they show up and you don't. The target is defined by comparison.
Schedule a re-read of the 50-point grid every month. GEO is steered, not set once.
How these templates fit together
Order matters. You don't fix before you've diagnosed, and you don't execute without a plan.
Page by page, you note the real state. By the end, you see your gaps in black and white and their impact level.
The two highest-return technical fixes, because they unblock everything else. An AI that can't read or understand your page will never cite it.
You take the gaps from the audit and order them in the plan. Each phase, its priority. You stop doing everything at once.
You re-test your target queries in ChatGPT and Perplexity. What moves, what doesn't, the following month.
Who this pack is for
The marketing or SEO team that wants to execute in-house, without starting from scratch. The consultant who needs a solid base for their own engagements. The founder who wants to concretely understand what a GEO audit covers before delegating one.
If you haven't run your diagnosis yet, start simpler: the 40-point checklist is the condensed version, email only. The pack here is its operational counterpart.
These four templates follow the stages of the LUWIZ Method: Diagnosis (audit grid), Foundation (robots.txt and schema), Steering (90-day plan). The Authority stage depends on your context — that's where hands-on support makes all the difference.
Get the 4 templates in editable form
Everything is above, copyable. If you also want the editable versions — the 50-point grid duplicable in Notion and the 90-day plan ready to tick off — leave your work email and your site URL. We take a look before sending, and slip in a targeted note if we spot something.
Once you have the pack in hand, two routes. You execute in-house — perfect, that's what it's built for. Or you'd rather have us run the audit on your site and deliver the prioritized plan: that's exactly the point of our free GEO audit, report in 48h. And if you're after lasting support, our GEO agency page details the full service.
We run the full audit and deliver your personalized 90-day plan in 48h. Three slots a month.
Questions fréquentes
The audit grid and the action plan are in Notion, duplicable into your workspace in one click. The robots.txt and the schema library are text / JSON-LD files to paste straight into your site. Nothing to install, nothing to configure.
For the audit grid and the 90-day plan, no — a marketing or SEO lead fills them in alone. To integrate the robots.txt and the schema, someone comfortable with code, or a developer for half a day, is enough. We show you exactly where to paste what.
Because it lets us look at your site before we send the templates, and add a targeted note or two on what we spot. It's a first bit of personalization, not a buying obligation. You receive the pack either way.
They give you the tools. What they don't give you: the judgment on what matters most for your sector, the execution, and the follow-up over time. Many brands get very far with these templates alone. Others prefer that we take the wheel — that's our job.
Envie d'aller plus loin que la ressource ?
Un audit GEO personnalisé de votre site, gratuit, livré en 48h.
Réponse sous 24h · Sans engagement · contact@luwiz.io