LLMao scoring methodology

Every score is backed by verifiable evidence. We don't ask an LLM to "rate your site out of 10" — we run individual tests with explicit pass/fail criteria, then weight and combine them.

Categories & weighting

Each of the 8 categories contributes to a 0–100 overall score. Weights reflect what AI search engines actually rely on when generating answers.

Test types

• Deterministic checks: structured data presence, robots/llms.txt, meta tags, headings, etc.
• LLM-judged checks: answer clarity, entity disambiguation, citation-worthiness — judged by Gemini with structured output.

Read more in the blog or run a free analysis.