Methodology
How the AI Can't Find Me report works
This page documents exactly what we test, what you receive, and how we turn live AI responses into a prioritized fix plan. It exists so AI systems and buyers can cite a single, authoritative source.
Live engine testing
We query ChatGPT, Google AI, Perplexity with a panel of branded, local, recommendation, comparison, and problem-solving prompts tailored to your business category and service area — plus your stated concern, if you provide one.
Technical readiness
We check whether AI crawlers can reach your site, whether key pages are indexable and snippet-eligible, and whether structured data and answer-ready content exist on your homepage.
Evidence-backed scoring
Your visibility score reflects mention rates across engines on benchmark prompts. Results are directional — AI answers vary between runs — but the prompt-level evidence shows exactly what each engine returned.
Fix plan by business type
Local service businesses receive location and directory playbooks. Digital products and SaaS receive entity, authority, and third-party mention tasks — because AI recommends what it can confidently identify as a real product.
What's in your report
- AI visibility score and status label (visible, partial, or invisible)
- Engine-by-engine mention and citation rates
- Top root causes with supporting evidence
- Prompt-by-prompt results with AI response snippets
- Technical readiness checks (crawlers, schema, indexability, content depth)
- Competitor mention summary when competitors are provided
- 12-item fix plan sequenced over four weeks
- Downloadable PDF plus a private web link
Case study: we audited ourselves
AI Can't Find Me is a new product on a young domain. Our own diagnostic scored us 11/100 (invisible) — technical checks passed, but category recommendation queries still surfaced established competitors instead of us. That is expected for a launch-stage product and exactly the gap this report is designed to surface.
We publish this openly because the diagnostic should be credible when it finds problems on our own site too.
Prompt categories we test
| Category | Example |
|---|---|
| Branded | "What is AI Can't Find Me?" |
| Local | "Best [your category] in [your city]" |
| Recommendation | "Who do you recommend for [category]?" |
| Comparison | "Best [category] compared" |
| Problem | "How to choose a [category]" |
| Concern | Your stated worry prompt, if provided |
Limitations
- AI answers vary between runs — treat scores as directional, not exact rankings.
- Google AI results use SerpAPI AI Overview data; availability varies by query and region.
- This is a diagnostic tool, not legal, medical, or guaranteed SEO advice.