How to read an AI Visibility Score, and what to fix first
Short answer
An AI Visibility Score is a 0–100 measure of how visible, trusted and recommended your brand is across AI engines, broken down across the five IEO pillars: Entity, Content, Authority, AI Accessibility and Distribution. Read the pillar scores, not just the total: the lowest pillar usually tells you what to fix first.
What the score is built from
In our free AI Visibility Audit, we run your brand through 20 buyer prompts across ChatGPT, Perplexity, Gemini and Google AI Overviews, and compare you with one competitor you choose. The results are scored against the five pillars of the IEO Framework™. Each pillar gets its own score, and together they make the 0–100 total.
What each pillar tells you
| Pillar | What it measures | A low score usually means |
|---|---|---|
| Entity | How consistently and accurately AI identifies who you are | Your description differs across your site, LinkedIn, directories and press |
| Content | Whether your pages give quotable answers to buyer questions | Answers are buried, pages describe features instead of answering questions |
| Authority | Whether trusted third parties mention and cite you | Little press, few reviews, no presence on the sources AI engines weight |
| AI Accessibility | Whether AI crawlers can read and understand your site | Blocked crawlers, missing structured data, content only visible after scripts run |
| Distribution | Whether your answers exist where AI learns | No presence in communities, video transcripts, podcasts or review sites |
How to read the citation map
The citation map shows every prompt in one of three states: you are cited, your competitor is cited, or neither is. Each state means something different:
- Competitor cited, you are not: the most urgent gap. The buyer is being sent elsewhere.
- Neither cited: an open opportunity. Nobody owns the answer yet.
- You are cited: protect it, and check the description is accurate.
Why accuracy matters as much as presence
Being mentioned with the wrong pricing, an outdated product or a competitor's feature can do more harm than not being mentioned. The accuracy check lists what AI engines get wrong about you today. Most errors trace back to an old source that engines still trust, which is why entity work often starts by correcting the record.
How to prioritise fixes
Every report ends with three fixes, ranked by impact and effort. As a rule of thumb:
- Fix AI Accessibility first if it is low. It is usually quick, and nothing else works while crawlers cannot read you.
- Fix Entity next. One consistent description is cheap and affects every prompt.
- Then work on Content for the prompts where competitors are cited, and Authority and Distribution for the long run.
