Case Studies
14.1 Case 1: A Small E-Commerce Store’s Three-Week Improvement
Background: A small-to-midsize outdoor gear e-commerce store with approximately 500 SKUs, running on WooCommerce, with a domain registered three years ago. Initial state:- Overall score: 52 (UNRATED)
- V: 55, S: 35, G: 40, T: 50, D: 30
- Had SSL but no DNSSEC, no DMARC
- No Schema.org Product markup
- No llms.txt
- Privacy policy was the default WordPress template
| Time | Action | Dimension Change |
|---|---|---|
| Day 1 | Enabled DNSSEC | S: 35 to 42 |
| Day 1 | Added DMARC + SPF | S: 42 to 52 |
| Day 2 | Installed RankMath, enabled Product markup | D: 30 to 45 |
| Day 3 | Created llms.txt | D: 45 to 52 |
| End of Week 1 | Rewrote privacy and return policies | T: 50 to 60 |
| Week 2 | Created agent.json | D: 52 to 55 |
| Week 2 | Added Organization markup to homepage | T: 60 to 64 |
| Week 3 | Optimized Sitemap to include all 500 products | D: 55 to 58 |
- The S dimension has the highest return on investment — two DNS records delivered a 17-point improvement
- D dimension Schema.org markup was auto-generated by a plugin, saving significant manual effort
- llms.txt took 15 minutes to write but boosted the D dimension by 7 points
14.2 Case 2: AI Visibility Optimization for a SaaS Product
Background: A B2B SaaS product providing project management tools, built on a Next.js custom site. Initial state:- Overall score: 48 (UNRATED)
- No structured data of any kind
- Had HTTPS but no DNS security configuration
- No llms.txt
- Policy pages existed but lacked Schema.org markup
- Organization markup — Structured company information
- SoftwareApplication markup — Structured product information
- FAQ markup — Structured frequently asked questions
- llms.txt — A concise description of product positioning, pricing, and core features
14.3 Case 3: The D-Dimension Gap for Major Brands
Analysis: An AI trust score analysis of well-known global brand domains reveals an interesting pattern. Many large brands score highly in V (identity verification) and G (governance) because they have GLEIF registrations, Wikidata entities, public company verification, and other credentials. However, they often underperform in the D (data quality) dimension. Root causes:- Large brand websites are typically managed by enterprise CMS platforms, and Schema.org markup is not always complete
- Large brands rarely have llms.txt
- Large brands’ robots.txt files may be overly restrictive, blocking AI crawlers
14.4 Optimization Efficiency Comparison by Platform
| Platform | Difficulty | Automation Level | Typical Time Investment |
|---|---|---|---|
| Shopify | Low | High (built-in Schema.org) | 2-3 hours |
| WooCommerce | Medium | Medium (plugins required) | 3-5 hours |
| Next.js / Custom | High | Low (development required) | 1-2 days |
| Wix / Squarespace | Low | Medium | 2-3 hours |
14.5 Common Failure Patterns
| Failure Reason | Symptom | Fix |
|---|---|---|
| Implemented but not validated | Schema.org markup has syntax errors | Use validation tools after every change |
| DNS records overwritten | DMARC records lost during domain migration | Check all DNS records immediately after migration |
| SSL certificate expired | Auto-renewal was not configured | Use auto-renewal (e.g., Let’s Encrypt) |
| robots.txt too restrictive | A newly installed security plugin blocked all crawlers | Check robots.txt after installing any plugin |
| llms.txt is stale | Product line has changed but llms.txt was not updated | Review once a month |
Congratulations on completing Book 6. You now have a comprehensive understanding of the AI-SEO knowledge framework. Recommended next steps:
- Put it into practice: Templates and Prompts — Ready-to-use configuration templates and AI prompts
- Dive deeper into trust scoring: Book 2: OTR Protocol
- Learn about commerce protocols: Book 3: UCP Protocol
- Look up terms: Glossary
More case studies: Getting Started Cases | OTR Cases | MCP Cases