AI Visibility

How should local visibility be measured when discovery occurs across search engines, maps, AI assistants, media surfaces and agents?

Visibility is per-surface, not a single ranking number, and each surface applies different retrieval rules.

Hypothesis — not a conclusion

Businesses cannot improve what they cannot measure, and the measure most of them use describes only one surface.

1. Record
   ↓
2. Surface
   ↓
3. Retrieval rule
   ↓
4. Appearance
   ↓
5. Measure

• BusinessRecordScore.com ↗
• LocalVisibilitySurfaces.com ↗
• TheVisibilityGrid.com ↗

• —That surfaces can be enumerated meaningfully
• —That a per-surface measure is workable in practice
• —That record quality correlates with appearance

A single ranking number no longer describes visibility across surfaces with different retrieval rules.

Testing has not yet produced documented results.

• Surface behaviour is too unstable to measure
• Record quality shows no relationship to appearance
• A single aggregate measure predicts outcomes just as well

• Can AI surfaces be sampled reliably enough to measure?
• Should the measure describe the record or the outcome?

Enumerate the surfaces and test whether a per-surface measure can be observed consistently.