AI audience building
Overview
Instead of building complex rule trees, you prompt the platform ("guests who visited more than twice in the last 60 days and haven't booked in the last 3 weeks"). The AI translates that to an audience; you refine and save it.
How it works
Natural-language prompts are mapped onto the schema of the unified guest profile (visit count, last booking, spend bands, loyalty tier, channel preferences). The resulting audience is live — profiles move in and out as their data changes.
Core concepts
AI prompt
free-text description of the audience you want.
Filter
explicit rule (e.g. `last_visit < 30 days`) for precise control.
Live audience
membership recalculates as profile data updates.
Usage & workflows
Drive campaigns, workflows, and exports. Audiences are also used as trigger conditions and exit criteria inside workflows.
Example
Build "at-risk regulars" (4+ visits in the last 12 months, none in the last 30 days) in one prompt, then attach a 15% win-back SMS workflow.
Want to see it live?
Book a 20-minute walkthrough and we'll show you the feature in your own admin.

