SCORING METHODOLOGY
How the AI Readiness score is calculated
The overall score (0-100) is a weighted sum of four evaluation axes. This page publishes what each axis measures and how.
Score composition
| Axis | Weight | What it measures |
|---|---|---|
| Text visibility | 45% | Match between browser-rendered text and text present in the AI-fetchable HTML |
| Image AI readability | 20% | Whether product images carry AI-readable descriptions |
| Information structure | 20% | Structured data and metadata implementation |
| Crawlability | 15% | Whether AI crawlers can access the page at all |
Text visibility carries the largest weight because body text is the primary material an AI uses to generate an answer. No matter how complete the structured data is, an AI cannot cite a page whose text it cannot fetch.
1. Text visibility (45%)
Method: the same URL is fetched two ways — a raw HTML fetch (how most AI crawlers request a page, without executing JavaScript) and a browser render (the fully JavaScript-executed HTML). The two bodies of text are compared; the larger the share of browser-visible text absent from the raw HTML, the greater the deduction. Sites built on SPA frameworks or JavaScript-heavy CMSs tend to score lowest here.
2. Image AI readability (20%)
Images that describe a product or service are checked for AI-readable text information (such as alt attributes).
Design decision: not every image counts. Logos, icons, and decorative images are excluded from scoring — only unexplained product images are penalized. A decorative image with no alt text is, in fact, correct HTML.
3. Information structure (20%)
- Presence and validity of structured data (JSON-LD, etc.)
- Basic metadata: title, meta description
- Logical heading structure
Structured data is the most reliable way to tell an AI what a page means — a product, an article, an organization — without ambiguity.
4. Crawlability (15%)
Whether the page can be fetched at all when accessed as an AI crawler (HTTP status, robots.txt control, etc.).
How “unavailable” is handled: if crawler access is blocked (HTTP 403, etc.), image and structure scores are not scored as zero — they are marked unavailable and excluded from the weighted average. A block is a configuration issue, not a content issue.
Limits and disclaimer
- This audit is a technical evaluation of publicly available information at the time of the check
- It does not guarantee search rankings, AI citation, traffic, or revenue
- robots.txt is respected; no changes are made to the audited site
See the disclaimer for details.