Skip to main content

​
Overview

The AI Insights Dashboard is your central hub for reviewing, filtering, and acting on AI-generated recommendations. Every AI feature in KaireonAI — policy recommendations, auto-segmentation, content intelligence, and rule suggestions — feeds its output into the Insights dashboard as discrete recommendation cards. Navigate to AI > Insights in the sidebar to open the dashboard.

​
Recommendation Lifecycle

Each recommendation follows a simple lifecycle:
StatusMeaning
NewFreshly generated, awaiting review
ReviewedOpened and read by a user
AppliedAccepted and converted into a draft entity
DismissedRejected — no further action
When a recommendation is generated, it starts as New. Opening the detail panel marks it Reviewed. From there you can either Apply it (which creates a draft entity in the appropriate module) or Dismiss it.

​
Filter Tabs

The dashboard provides filter tabs to narrow the recommendation list:
  • All — Every recommendation across all AI features
  • Policy — Contact policy optimization suggestions from the Smart Policy Recommender
  • Segmentation — Customer segment discoveries from Auto-Segmentation
  • Content — Creative performance insights and rewrite suggestions from Content Intelligence
  • Rules — Qualification rule suggestions generated via the AI chat panel
Each tab shows a count badge indicating how many New recommendations are available.

​
Recommendation Cards

Each card displays:
  • Title — A concise summary of the recommendation (e.g., “Reduce email frequency for segment High-Value”)
  • Type badge — Which AI feature generated it (Policy, Segmentation, Content, Rules)
  • Confidence — How confident the AI is in the recommendation (High, Medium, Low)
  • Impact — Estimated business impact (High, Medium, Low)
  • Created at — When the recommendation was generated
  • Status — Current lifecycle status
Click a card to expand the detail panel, which shows the full analysis, supporting data, and action buttons.

​
Applying Recommendations

When you click Apply on a recommendation:
  1. KaireonAI creates a draft entity in the appropriate module — for example, a draft Contact Policy, a draft Qualification Rule, or a draft Creative variant
  2. You are redirected to the relevant editor to review and finalize the draft
  3. The recommendation status updates to Applied with a link to the created entity
Draft entities are never auto-activated. You always have the opportunity to review, modify, and explicitly activate them.

​
ML Worker Status

The dashboard header shows the ML Worker connection status:
  • Connected (green) — The ML Worker is running and available for sklearn/TensorFlow analysis
  • LLM Only (amber) — No ML Worker detected; AI features fall back to LLM-based analysis
  • Disconnected (red) — Neither ML Worker nor LLM provider is configured
When the ML Worker is connected, AI features that support dual-tier routing (like Auto-Segmentation and Smart Policy Recommender) automatically use the ML Worker for large datasets, producing more accurate results.
You can configure the ML Worker connection in Settings > Integrations > ML Worker. See the ML Worker Setup guide for deployment instructions.

​
Next Steps

Smart Policy Recommender

AI-powered contact policy optimization.

Natural Language Rule Building

Create rules by describing them in plain English via the AI chat panel.

Auto-Segmentation

Discover customer segments from your data.