AI Visibility Score

AI Visibility Score is a 0 to 100 score that summarizes how visible you are for a specific query.

It is designed to be:

• Deterministic
  Same inputs always produce the same score

• Explainable
  You can see which components contributed

• Auditable
  Scores are stored as append only snapshots

What the score measures

AI Visibility Score combines multiple signals:

1. Retest performance
   Where you rank for the query in the latest retest run

2. Schema coverage
   Whether you have completed the relevant schema improvements

3. Content freshness
   How recently you have refreshed or approved content tied to that query

Each component produces a 0 to 100 value, then a weighted average is computed.

If a component is missing, weights are renormalized so the score still makes sense.

What it is not

AI Visibility Score is not:

• A guaranteed predictor of revenue
• A replacement for analytics like GA4
• A promise that rankings will increase after a change

It is a directional score for tracking progress over time on a per query basis.

Where you will see it

You will usually see AI Visibility Score in:

• Dashboard summary
  A snapshot of current health

• Query detail views
  Score history over time for a specific query target

• Retest history
  New snapshots created after retests complete

How score history works

Scores are stored as snapshots. This means:

• Past scores never change
• You can compare score trends week to week
• You can audit what inputs were used for each snapshot

This makes the score trustworthy for reporting and decision making.

How to improve the score

There are three levers.

Improve retest performance

• Make sure the query target has a clear page that matches intent
• Use Optimize Result outputs to strengthen content for that query
• Address technical issues that block indexing or clarity

Improve schema coverage

• Run and approve schema tasks for the surfaces relevant to the query
• Keep schema accurate and aligned with store catalog

Improve content freshness

• Refresh approved content for the query periodically
• Use brand voice settings to make outputs easier to publish
• Avoid letting high value pages go stale

Common patterns

Score drops suddenly

Common causes:

• A rank position drop in retest results
• A missing signal due to an API quota or rate limit
• Content freshness aging past a threshold

If the drop is large, the system can create follow up tasks for review.

Score changes but rankings look stable

This can happen when:

• Schema coverage changed after completing tasks
• Content freshness changed due to time since last update
• One component gained or lost data, causing weight renormalization

Troubleshooting

If you see missing data or unexpected gaps:

• Check whether a retest run completed successfully
• Check plan quotas and rate limit messages
• Review errors in Troubleshooting then Common errors

Related pages

• Features then Optimize Result
• Core concepts then Tasks
• Workflows then Preview approve execute rollback
• Troubleshooting then Common errors
• Glossary then Terms