> ## Documentation Index
> Fetch the complete documentation index at: https://learn.orbexa.io/llms.txt
> Use this file to discover all available pages before exploring further.

# AI commerce real-world case studies

> AI trust score case studies with publicly verifiable data from over 3,200,000+ domains — real merchant scores, optimization paths, and results.

# Real-World Case Studies

The following cases are based on publicly verifiable data analysis of over 3,200,000+ domains. You can use AI trust assessment tools such as ORBEXA ([orbexa.io/verify](https://orbexa.io/verify)) to look up the trust score of any domain.

## Case 1: Trust Score Distribution Among Major Brands

We analyzed the OTR trust scores of well-known global brands. One surprising finding: **big brands do not automatically score high.**

| Dimension        | What High-Scoring Brands Have in Common                   | Where Low-Scoring Brands Fall Short                   |
| ---------------- | --------------------------------------------------------- | ----------------------------------------------------- |
| V (Verification) | Complete SSL, GLEIF registration, Wikidata entity         | Basic SSL only, no business registration verification |
| S (Security)     | DNSSEC + DMARC + SPF + DKIM all configured                | SSL only, DNS security missing                        |
| G (Governance)   | LEI number, publicly verifiable listing                   | No legal entity registration                          |
| T (Transparency) | Complete privacy/return policies, Schema.org Organization | Policies vague or missing                             |
| D (Data Quality) | JSON-LD Product markup + llms.txt + Sitemap               | No structured data at all                             |

**Key finding**: The D dimension (Data Quality) is the easiest to improve. Many major brands score well on V/S/G, but their D dimension drags down the total because they lack llms.txt and complete Schema.org markup.

## Case 2: From BRONZE to SILVER -- A Score Improvement Path

A typical small-to-midsize e-commerce site started at a score of 52 (UNRATED). Through the following steps, it reached 72 (SILVER):

**Week 1: Security Foundations (+8 points)**

* Configured DNSSEC -> S dimension improved
* Added DMARC record -> S dimension improved
* Added SPF and DKIM -> S dimension improved

**Week 2: Transparency (+6 points)**

* Rewrote privacy policy (from a generic template to a detailed, custom version) -> T dimension improved
* Added return policy details -> T dimension improved
* Added complete contact information -> T dimension improved

**Week 3: Data Quality (+6 points)**

* Added Schema.org Product markup to all product pages -> D dimension improved
* Created llms.txt -> D dimension improved
* Optimized Product Sitemap -> D dimension improved

**Total improvement: approximately 20 points, from 52 to 72, jumping from UNRATED to SILVER.**

<Tip>
  None of this work requires paid tools. DNSSEC can be enabled in your domain registrar's dashboard. DMARC/SPF/DKIM are DNS records you add for free. Schema.org markup is HTML code. The llms.txt file is plain text. All of it is free.
</Tip>

## Case 3: The D Dimension Is the Biggest Score Lever

Based on analysis of global e-commerce domains:

* Fewer than 5% of domains have an llms.txt file
* Fewer than 30% of e-commerce sites have complete Schema.org Product markup
* Fewer than 3% have both

This means: **if you do what others have not, you are ahead of 97% of your competitors.**

Every point in the D dimension tells AI agents: "This merchant's data is structured, complete, and machine-readable." AI agents will prioritize merchants with high data quality because they can more accurately understand and recommend those products.

## Your Action Plan

Ranked by priority -- start with what makes the biggest impact:

### Do Today (30 minutes)

* [ ] Check your trust score -> [orbexa.io/verify](https://orbexa.io/verify)
* [ ] Create an llms.txt file -> place it in your website's root directory
* [ ] Check robots.txt -> confirm AI crawlers are not blocked

### Do This Week (2-3 hours)

* [ ] Add Schema.org Product markup to product pages
* [ ] Configure a DMARC record
* [ ] Configure an SPF record
* [ ] Review and optimize your Sitemap

### Do This Month (1-2 days)

* [ ] Rewrite your privacy policy and return policy
* [ ] Enable DNSSEC
* [ ] Add Schema.org Organization markup
* [ ] Create an agent.json file

### Long-Term Goals

* [ ] Target trust score of 70+ (SILVER)
* [ ] Adopt the UCP protocol
* [ ] Adopt the ACP protocol (requires developer involvement)

***

Congratulations on finishing Book 1. You now understand the full landscape of AI agent commerce.

**Recommended next steps:**

* Ready to implement -> [SEO for AI](/en/book-6/index)
* Want a deeper understanding of trust scoring -> [OTR -- Open Trust Registry](/en/book-2/index)
* Look up a term -> [Glossary](/en/glossary/index)

***

**Recommended next read**: [OTR -- Open Trust Registry](/en/book-2/index) -- a deep dive into the six-dimension trust assessment system

**More case studies**: [OTR Cases](/en/book-2/ch12-case-studies) | [UCP Cases](/en/book-3/ch10-case-studies) | [SEO Cases](/en/book-6/ch14-case-studies)