What AI Pilot uses
Only the selected client's own context:
- data from connected sources (Search Console, Google Analytics 4 and connected ad accounts);
- the latest completed Growth Audit;
- the latest saved competitor analysis (qualitative);
- the Brand Profile.
It does not browse the web live for your answer.
How answers are labelled
- Based on N measured sources, or AI judgement when no measured data was cited.
- AI interpretation: the answer is an interpretation, not a measurement.
- High, Medium or Low confidence.
- Evidence used lists each source with its label (Measured, User-provided, Connected data or AI-discovered) and freshness (Fresh, Ageing, Stale or Not synced yet).
- Limitations and assumptions is always shown.
If an answer cites no data, it says: "Not enough data. This answer is AI judgement only… Treat it as general guidance, not a finding."
Limitations
AI Pilot can be wrong. It is only as good as the evidence available, and stale data makes answers less reliable. Check the evidence before acting.