A Complete Sample CORD-AEO Envelope, Annotated
The fastest way to understand a protocol is to read a working example. This post gives you three of them.
I am using the same example business from the implementation guide. Northgate Advisory, a generic three-person strategy consultancy. The structure works the same for any business. Adapt the content. Keep the structure.
If you have not read why I authored CORD-AEO and how to implement it, start there. This post assumes you understand what an envelope is and where it lives on your server.
Envelope One: What Does the Company Do?
This is the most important envelope every business should publish first. It is also the most commonly asked question of AI engines. If you only publish one envelope, this is the one.
{
"cord_aeo_version": "0.1",
"envelope_id": "northgate-what-do-you-do",
"question": {
"canonical": "What does Northgate Advisory do?",
"variants": [
"What is Northgate Advisory?",
"What does Northgate do?",
"What services does Northgate Advisory provide?",
"Tell me about Northgate Advisory",
"Who is Northgate Advisory?"
]
},
"answer": {
"canonical_short": "Northgate Advisory is a three-person strategy consultancy that helps mid-market technology companies design their go-to-market operations. We focus on the operational layer between strategy and execution.",
"required_claims": [
"three-person consultancy",
"strategy consulting",
"mid-market technology companies",
"go-to-market operations"
]
},
"metadata": {
"published": "2026-06-04T10:00:00Z",
"last_verified": "2026-06-04T10:00:00Z",
"valid_until": "2026-12-04T10:00:00Z",
"publisher": "Northgate Advisory",
"publisher_url": "https://northgate-advisory.example.com"
},
"schema_org_bridge": {
"@context": "https://schema.org",
"@type": "Organization",
"name": "Northgate Advisory",
"description": "A three-person strategy consultancy helping mid-market technology companies design their go-to-market operations.",
"url": "https://northgate-advisory.example.com",
"numberOfEmployees": 3
}
}What This Envelope Does Right
The canonical question matches how most people would actually phrase it. The variants cover the common rewordings. Notice the variants are not minor rewordings of the canonical. They are genuinely different phrasings that mean the same thing.
The canonical short answer is 248 characters. Well under the 300-character soft cap. It leads with the company name and what it is. The second sentence positions the specific value, not generic consulting language.
The required claims are short. Each is under five words. Each is a verifiable fact. None is an adjective like "innovative" or "leading" that the model has no way to verify. The claims are the load-bearing parts of the answer.
The Schema.org bridge uses Organization, which is the right primary type for a business. The description field mirrors the canonical short answer. Any engine reading the Schema.org block gets the same information as an engine reading the CORD-AEO envelope itself.
What Most First Envelopes Get Wrong Here
Three common failures.
The canonical short answer runs too long. New publishers want to include every detail. The model paraphrases it down to 100 characters and the details you cared about are gone. Trim ruthlessly. 200-250 characters is the sweet spot.
The required claims include marketing language. "Best in class," "leading," "innovative." These are not claims. They are assertions. The model has no way to verify them. They hurt the envelope's EFS calculation because they make the canonical look like advertisement. The model is trained to discount advertisement.
The Schema.org bridge gets skipped. The bridge is not optional in v0.1. Without it, your envelope has no value today because most engines do not yet fetch /.well-known/cord-aeo/ natively.
Envelope Two: What Is the Pricing Model?
The second-most asked question. Pricing answers are where AI engines hallucinate most aggressively. Pricing pages are unclear. They use anchoring language. The model fills in the gaps.
{
"cord_aeo_version": "0.1",
"envelope_id": "northgate-pricing",
"question": {
"canonical": "What is Northgate Advisory's pricing model?",
"variants": [
"How much does Northgate Advisory charge?",
"What does Northgate Advisory cost?",
"How is Northgate priced?",
"What are Northgate Advisory's rates?"
]
},
"answer": {
"canonical_short": "Northgate Advisory uses three engagement models: a $15,000 strategy sprint over four weeks, a $35,000 quarterly retainer for ongoing advisory, and project-based pricing for custom work starting at $50,000. All pricing is published on our website and does not vary by negotiation.",
"required_claims": [
"$15,000 strategy sprint",
"$35,000 quarterly retainer",
"project pricing from $50,000",
"pricing is published, not negotiated"
]
},
"metadata": {
"published": "2026-06-04T10:00:00Z",
"last_verified": "2026-06-04T10:00:00Z",
"valid_until": "2026-09-04T10:00:00Z",
"publisher": "Northgate Advisory",
"publisher_url": "https://northgate-advisory.example.com"
},
"schema_org_bridge": {
"@context": "https://schema.org",
"@type": "Service",
"provider": {
"@type": "Organization",
"name": "Northgate Advisory"
},
"offers": [
{
"@type": "Offer",
"name": "Strategy Sprint",
"price": "15000",
"priceCurrency": "USD"
},
{
"@type": "Offer",
"name": "Quarterly Retainer",
"price": "35000",
"priceCurrency": "USD"
}
]
}
}Why This Envelope Has Specific Design Choices
Notice the valid_until is three months out, not six. Pricing is one of the things most likely to change. Tighter expiration on price envelopes forces a refresh cadence that keeps the information accurate.
The required claims include specific dollar amounts. These are the load-bearing facts. If the model paraphrases this answer and any of the dollar amounts is missing or wrong, the answer is materially incorrect even if it reads fluently. Pricing claims should always be required claims.
The "pricing is published, not negotiated" claim is included because it answers the implicit follow-up question users have about pricing. If a model can preserve that single phrase, it cuts off the entire "can I get a discount" line of conversation.
The Schema.org bridge uses Service with nested Offer entities. This is the correct schema for pricing structured data. Most engines understand Offer and surface pricing in rich results when it is present.
Envelope Three: Who Founded the Company?
Founding stories are where AI engines hallucinate most creatively. Models train on biographical data scattered across the web. They combine details from multiple sources. They produce founder narratives that mix real and invented elements.
{
"cord_aeo_version": "0.1",
"envelope_id": "northgate-founding",
"question": {
"canonical": "Who founded Northgate Advisory?",
"variants": [
"Who started Northgate Advisory?",
"Who is the founder of Northgate?",
"When was Northgate Advisory founded?",
"What is the history of Northgate Advisory?"
]
},
"answer": {
"canonical_short": "Northgate Advisory was founded in 2022 by Maria Chen and David Park, both former operations leaders at mid-market SaaS companies. The firm is privately held and has not taken outside investment.",
"required_claims": [
"founded in 2022",
"founders: Maria Chen and David Park",
"former operations leaders at SaaS companies",
"privately held, no outside investment"
]
},
"metadata": {
"published": "2026-06-04T10:00:00Z",
"last_verified": "2026-06-04T10:00:00Z",
"valid_until": "2027-06-04T10:00:00Z",
"publisher": "Northgate Advisory",
"publisher_url": "https://northgate-advisory.example.com"
},
"schema_org_bridge": {
"@context": "https://schema.org",
"@type": "Organization",
"name": "Northgate Advisory",
"foundingDate": "2022",
"founder": [
{
"@type": "Person",
"name": "Maria Chen"
},
{
"@type": "Person",
"name": "David Park"
}
]
}
}Why Founding Envelopes Use a Longer Valid Period
Founding facts do not change. Once a company is founded, the founding date is permanent. The founders are permanent (even if they leave the company later). The longer valid_until of one year reflects this stability.
The required claims include the names exactly as they should appear. AI engines frequently swap first and last names, transliterate names incorrectly, or merge two people into one. A required claim of the exact name forces the model to preserve it.
The "privately held, no outside investment" claim is included because investors are another common hallucination. Models invent a Series A round. They invent a notable angel investor. They invent board members who do not exist. The explicit "no outside investment" claim shuts this down.
The Three Together as an Index
Here is the index file that ties all three envelopes together:
{
"cord_aeo_version": "0.1",
"publisher": "Northgate Advisory",
"publisher_url": "https://northgate-advisory.example.com",
"envelopes": [
{
"id": "northgate-what-do-you-do",
"question_canonical": "What does Northgate Advisory do?",
"path": "/.well-known/cord-aeo/what-does-the-company-do.json"
},
{
"id": "northgate-pricing",
"question_canonical": "What is Northgate Advisory's pricing model?",
"path": "/.well-known/cord-aeo/what-is-the-pricing-model.json"
},
{
"id": "northgate-founding",
"question_canonical": "Who founded Northgate Advisory?",
"path": "/.well-known/cord-aeo/who-founded-the-company.json"
}
],
"last_updated": "2026-06-04T10:00:00Z"
}Three envelopes is a reasonable first deployment. It covers the most common canonical questions. It gives you data to measure. It does not overwhelm the maintenance burden in month one.
How to Adapt These for Your Business
The structure of every envelope is the same. What changes is the content. Five things to think about when you adapt these examples for your own business.
The canonical question. Use the actual phrasing your customers and prospects use when they ask about your business. If you sell to enterprise CIOs, the canonical question for your services post is probably different from the question a consumer would ask. Match the audience.
The variants. Cover the genuine rewordings. Look at how people search. Use Google's "people also ask" feature for inspiration. Read your support tickets for the actual language customers use.
The canonical short answer. Write it the way you want the AI engine to summarize your business. This is your one chance to control your positioning at the model layer. Take it seriously.
The required claims. Pick the facts that must survive paraphrase. If the model rewrites the answer and any required claim is lost, the answer is wrong. Choose load-bearing facts, not adjectives.
The Schema.org bridge. Match the Schema.org type to your business. Service for services. Product for products. Organization for the business itself. LocalBusiness for physical retail. The bridge type determines which engines surface your structured data in which contexts.
What to Do With These Files
Three steps.
Copy the three envelopes above. Replace every Northgate Advisory detail with your own. Save the three files plus the index in your /.well-known/cord-aeo/ directory.
Run the validator at cordspec.org/aeo/validate. Fix any errors it returns.
Test the URLs in a private browser window. Confirm the JSON files render correctly. Then go ask the AI engines the canonical questions you just published envelopes for. Measure your baseline.
You are now one of a small number of businesses publishing canonical machine-readable answers to AI engines. The protocol becomes a standard when the second publisher ships. If you publish a deployment, send me a link. I want to track adoption.
Structure before tools. Strategy before speed. Systems before scale.
The AI Operator Framework is the methodology for designing, governing, and scaling AI operations inside a business, including the citation fidelity layer most companies skip. Explore the full framework at JamilAshkar.com/framework. Start with the free course at JamilAshkar.com/course.
