AI Health
The whole picture, before you go looking at the parts.
Evidence arrives from a lot of places at once. AI Health pulls it into a single readable state across four pillars, so you can tell where the weakness actually sits before spending a morning in the detail.
No card required.
- Four readiness pillars
- Sub-scores, not just a headline
- Re-checked as work lands
Four pillars
Readiness is not one thing, so it is not measured as one thing.
Each pillar can fail independently, and a problem in one is rarely fixed by work in another. Keeping them separate is what makes the summary useful rather than merely reassuring.
- 01
Access
Can it be reached at all?
Whether crawlers and agents can get to your pages in the first place. Nothing downstream matters if this is failing.
- 02
Indexability
Is it eligible to be used?
Whether what can be reached is actually available to be included, rather than excluded by directive or configuration.
- 03
Extractability
Can the content be parsed?
Whether the substance of a page can be pulled out cleanly, rather than being locked inside a structure that obscures it.
- 04
Answerability
Does it answer anything?
Whether the content actually addresses the questions being asked, in a form that can be drawn on to build a response.
How to read it
The headline is an entry point, not a verdict.
A single summary figure is useful for knowing whether to look closer. It is a poor tool for deciding you are fine.
-
A strong overall reading can hide a hard failure
Three healthy pillars and one broken one can still average into something comfortable. The breakdown is where the actual information is.
-
Pillars are not interchangeable
An access problem is not offset by strong answerability. The weakest pillar tends to set the ceiling for everything above it.
-
It describes evidence, not outcomes
Readiness reflects the state of your site as observed. It is not a forecast of how answer engines will behave.
Movement
Useful mainly as something you watch, not something you read once.
Readiness is re-checked as work lands, so the value builds up over time. A pillar recovering after a Case closes is a far more meaningful signal than any single reading taken in isolation, and it is the version that belongs in a report.
Readings depend on underlying scan data, so re-run the relevant evidence after making changes rather than expecting the summary to move on its own.
What AI Health connects to
The summary is assembled from evidence held elsewhere, and hands off to the places where work happens.
See where your site actually stands.
Run an evidence pass and get the readiness picture across all four pillars.
Want a read on your current state? Talk to us