AI Visibility Readiness Score.
Ten checks, two points each. Score what is true today—not what is on the roadmap. This measures readiness, not current rankings or guaranteed citations.
How to score
Give yourself 2 points when the statement is consistently true, 1 point when it is partly true, and 0 points when it is unknown or false.
- 01Public access
Priority pages return useful
200HTML without a login, challenge or client-only content dependency. - 02Crawler policy
Robots, page-level directives and the CDN deliberately allow the search crawlers you want.
- 03Discovery
Priority pages have crawlable internal links, correct canonicals and an accurate XML sitemap.
- 04Direct answers
Pages state definitions, facts, comparisons and limitations clearly, with the important answer visible early.
- 05Original value
Your site publishes first-hand expertise, data, methods or decision guidance that generic summaries do not.
- 06Entity consistency
Name, category, descriptions, product facts and other core information agree across reputable sources.
- 07Independent corroboration
Relevant third parties discuss or validate your product, expertise or claims.
- 08Prompt map
You maintain a prioritized list of real questions by audience, intent and buying stage.
- 09Repeatable measurement
You rerun a stable sample across relevant engines and track mentions, citations, accuracy and competitors.
- 10Business attribution
You can identify AI referrals or otherwise connect visibility changes to qualified outcomes.
Fix eligibility first.
Resolve access, discovery and basic information problems before investing heavily in content volume or outreach.
Build a baseline.
Some pieces exist, but measurement and execution are not connected. Start with technical checks and a prompt/source audit.
Prioritize gaps.
Your foundation is credible. Diagnose where specific competitors or sources win and ship focused improvements.
Iterate and attribute.
You have a repeatable program. Improve evidence, authority and connection to business outcomes.
Want an observed score instead?
An audit tests actual prompts, citations, sources and technical responses rather than relying on self-reporting.