How many prompts do I need for a statistically meaningful audit?

30 prompts per intent bucket run daily for 30 days gives ~900 response observations per bucket — enough to surface stable Visibility % trends. Below 100 total prompts, weekly noise can drown the signal.

Do I need to track every AI surface?

No — match the platform list to where your buyers actually are. For most B2B categories, that's ChatGPT + Perplexity + Claude + Gemini at minimum. Add Google AI Overview if you care about Google search intent collapsing into AI.

How does Astiva define brand normalization?

The normalization engine resolves mentions across casing, spacing, hyphenation, common misspellings, and brand aliases — cross-validated against human-labeled ground-truth. See the methodology page for the reporting windows and the full normalization spec.

How to Audit Your AI Visibility

A defensible AI visibility audit uses 7 metrics, a published methodology, and at least 30 days of daily data. This framework is the one Astiva ships to customers.

The 7 AISO metrics

Visibility % — mention rate across tracked prompts
Share of Voice — brand mentions vs. competitor mentions
Average Position — ordinal position within responses
Sentiment — polarity of each mention
First Mention Rate — share of responses where brand is named first
Mention Frequency — mentions per response when present
Sentiment Volatility — week-over-week variance in sentiment

See the full methodology with formulas and reporting windows.

← Back to Astiva AI