/ Methodology version v2026.10 · Last updated October 23, 2026
Our methodology
How we build the AI Visibility Readiness Score — and what it can and can’t tell you.
We believe a score is only useful if you can understand it. That’s why we explain what we measure, how we calculate it, and where its limits are.
Readiness, not rankings.
The AI Visibility Readiness Score is a 0–100 assessment of how ready a business website is to be understood, trusted, cited and recommended by answer engines. It measures the signals on your website that you can control. It does not measure — or predict — your position in any specific AI tool. It doesn’t certify the quality of your services.
/ Score labels
Find your score on the scale.
Positive signals, along with important gaps worth prioritizing.
The site may be hard for AI systems to understand, trust or recommend.
Positive signals, along with important gaps worth prioritizing.
A useful foundation, with some gaps in detail, evidence or structure.
Solid signals for being understood and evaluated by AI systems.
We couldn’t extract enough reliable content to give a definitive score.
/ The seven categories
Where your 100 points come from.
1 square = 1 point of the AI Visibility Readiness Score
Service Clarity carries the most weight because everything else depends on it: if your offer isn’t clear, nothing else can make up for it.
From signals to a score.
- Weighted score. Each category is scored and weighted to produce an overall score.
- Critical gates. Some gaps are serious enough to limit the overall result, regardless of how strong other categories are.
- Label. The final score is given a label that explains what it means.
Why some gaps limit your score.
A high average can hide a fundamental problem. Critical gates prevent that.
- If Service Clarity is below 50, the score is capped at 64 (Opportunity to Improve). If your offer isn’t clear, AI can’t understand or recommend you with confidence.
- If Trust Signals are below 50, the score can’t be labeled Strong AI Visibility Signals. Without enough evidence, a strong recommendation isn’t defensible.
- If a local business has Local Signals below 50, the score is capped at Opportunity to Improve. For local businesses, location and service area are essential.
- If your robots.txt verifiably blocks all three major AI search crawlers (OAI-SearchBot, Claude-SearchBot and PerplexityBot), the score is capped at 79. If AI search tools can’t access your site, they can’t cite it. Blocking AI training crawlers doesn’t count, and neither does a robots.txt we can’t verify.
- If we can’t reliably read your site — for example, because content loads only through scripts or access is blocked — we show Needs Manual Review instead of a score that might be wrong.
When a gate applies, your report tells you which one and why.
/ Try it
Build a score yourself.
Good
A useful foundation, with some gaps in detail, evidence or structure.
This is a simplified model. Real audits score dozens of checks inside each category, and a page we can't read reliably gets Needs Manual Review instead of a score.
Evidence first.
Every finding follows the same structure: what we observed, what it means for your business, the evidence behind it, and the recommended action. Each finding links to at least one recommendation, so nothing is left as an unexplained judgment.
What we use, and what we don’t.
- We analyze the public page you submit, plus your domain’s robots.txt and AI bot access to that page.
- We identify ourselves as VisiblaBot/1.0.
- We check robots.txt for every page and redirect. If it blocks us, the audit stops.
- We don’t bypass logins, access errors, anti-bot challenges or other technical controls.
- We don’t use private data or internal systems.
- Page content is processed transiently. Summarized signals may be cached for up to 30 days.
- Each report is a snapshot of your site on the date of the analysis.
- The score reflects readiness signals. It doesn’t measure traffic, revenue or results in any specific AI tool.
- Other pages, such as individual service pages or articles, may get different results.
A score that evolves — transparently.
Answer engines change, and our methodology will too. Every report states the methodology version used. When we update weights, labels or critical gates, we release a new version. Reports created under an earlier version keep their original score. New audits use the current version, so we’ll always tell you when a score change comes from the methodology rather than from your website.
View methodology changelog →How we keep the score honest.
We test the score against a reference set of strong, average and weak websites in each industry, and we manually review a sample of audits, especially paid reports and unusual results. If you believe your score is wrong, contact us. We’ll review the evidence, how reliably your site was read, and the signals behind each category. If we find an error, we’ll correct the report and record why.
Request a score review →Disclaimer
The AI Visibility Readiness Score is an assessment of readiness signals. It is not a guarantee of visibility, mentions, citations, recommendations or business results in any AI system. VISIBLA is not affiliated with OpenAI, Google, Anthropic, Perplexity, xAI or any other AI provider.
See where your website stands.
Frequently asked questions
Does a high score guarantee that AI will recommend my business?
No. It measures readiness. No one can honestly guarantee what an AI will say. The score reflects public signals on your website that help AI systems understand and evaluate your business, and it shows you what to improve first. It is not a guarantee of rankings, traffic, mentions or recommendations.
Why is my score capped?
Because a critical gate applies. Some gaps are serious enough to limit the overall score, however strong other categories are: for example, an unclear offer, weak trust signals, or blocking AI search crawlers. Your report tells you which gate applies and why. See critical gates →
Can my score change if I haven’t changed my website?
It can, if our methodology changes. Every report states the methodology version used. Earlier reports keep their original score, and we’ll always tell you when a change comes from the methodology rather than from your website.
What if I think my score is wrong?
Contact us. We’ll review the evidence, how reliably your site was read, and the signals behind each category. If we find an error, we’ll correct the report.
Why couldn’t my website be analyzed?
Usually because access is restricted. Your robots.txt may block VisiblaBot, the page may require a login or use anti-bot protection, or the content may load only through scripts. We never bypass those controls. When we can’t read a page reliably, we show Needs Manual Review instead of a score that might be wrong.