Skip to main content
The dashboard is the first screen you see after logging in. It gives you an at-a-glance view of sessions assigned to you and the overall state of your organization.

Screen layout

The dashboard is organized into the following sections. Dashboard overview screenshot

Hero section (top)

MetricDescription
SessionsTotal number of currently active sessions
ResolvedNumber of sessions resolved by the AI or an operator
EscalatedNumber of sessions currently escalated to a human operator
User FeedbackNumber of feedback items received
The “AI Working” count shows how many sessions the AI is currently handling. A higher number means the AI is actively responding to more requests.

3-column bento grid (middle)

Status distribution

A donut chart showing the breakdown of all sessions by status. Hover over each segment to see details.

Resolution rate

AI-resolved sessions ÷ total sessions as a percentage. Green at 60%+, yellow at 30–60%, red below 30%.

User Feedback

The number of feedback items received from users.

Status types

StatusColorMeaning
AI WorkingBlueAI agent is handling the session
OpenYellowWaiting for a response
EscalatedRedEscalated to a human operator
Awaiting UserOrangeWaiting for the user to reply
ResolvedGreenIssue resolved

To Do list (bottom)

A list of sessions assigned to you that need attention.
  • Escalated (red): Sessions escalated to you that require human action
  • Assigned (blue): Sessions assigned to you
Click Respond now on any row to open the ticket detail.

Org Overview (admins only)

Admins see an Org Overview section with summary cards for all sessions in the organization. Cards are broken down by the five status types.
CardContent
OpenSessions waiting for a response
AI WorkingSessions the AI is handling
Awaiting UserSessions waiting for the user
EscalatedSessions escalated to an operator
ResolvedResolved sessions
Org Overview is only visible to Admins. Operators do not see this section.

Improving resolution rate

If the resolution rate is low, consider the following:
  • Whether the AI has enough knowledge to answer questions (knowledge base coverage)
  • Whether escalation patterns are concentrated in specific inquiry categories
  • Whether any common inquiries could be handled automatically with playbooks
For detailed data analysis, refer to the tickets and analytics pages.
Last modified on July 10, 2026