Technical layer
5 min readAI Readiness Scans
Before any content strategy matters, AI engines have to be able to fetch, render and parse your site. The readiness scan runs 30+ technical checks against your domain and returns a 0-100 score with a letter grade and a fix for every failed check. It runs against your own site only and completes in under a minute.
What the scan checks
- Crawler access: Whether GPTBot, ClaudeBot, PerplexityBot and other AI crawlers can reach your pages, or are blocked by robots.txt, CDN and WAF rules, the most common silent failure.
- Rendering: Whether your content is readable without executing JavaScript. Many AI crawlers do not run your scripts; a JS-only page can be invisible to them while looking fine in a browser.
- Structured data: Organization, product and FAQ schema that lets engines identify your entity and lift facts with confidence.
- Answerability: Metadata, headings and content structure signals that make pages liftable into answers, plus llms.txt presence.
Reading the score
Every check reports pass or fail with a plain-language explanation and the specific fix. The score weights the checks by how much they block engines in practice: an A with two cosmetic warnings and an F with a CDN-level bot block are very different situations even if both have items to fix.
Treat readiness as a gate, not a goal. A strong score does not create recommendations; it removes the ceiling on them. Once you pass the technical layer, share of voice work is where visibility is actually won.
The failures we see most
- Bot protection blocking AI crawlers: CDN or WAF rules that challenge or reject GPTBot and PerplexityBot while letting Googlebot through. Usually one allowlist rule to fix, and worth checking after every infrastructure change.
- Client-side rendering: Key pages that render empty without JavaScript. Fix with server-side rendering or prerendering for your marketing pages at minimum.
- Missing entity schema: No Organization markup, inconsistent naming across pages, or no disambiguation from similarly named companies. Engines hesitate to recommend brands they cannot confidently identify.
Field note
The most expensive readiness failures are the ones introduced after the audit passed: a CDN migration, a new bot-protection vendor, a frontend rewrite. Re-scan after every infrastructure change, not just after content changes.
Re-scanning after fixes
Run a fresh scan after each batch of fixes ships. Scans are quota-friendly by design and do not consume audit credits: audits measure answers, scans measure your site. Starter includes 8 scans per month, Pro includes 24, Agency Pro includes 40 pooled across client projects.
There is also a free public version of the checker at citedintel.com/ai-seo-checker that runs without an account, limited to 2 checks per day per visitor. It is the same class of checks, useful for a first look or for checking a prospect's site before a pitch.