Live dashboard

The Dashboard is the supervisor’s real-time operations view. It opens at Dashboard in the sidebar and refreshes every 30 seconds, plus a manual Refresh button in the top-right of the page header. There is no live socket push today — the 30-second poll is what keeps tiles current, and a small Live badge next to “vs. yesterday” turns to Stale if the latest snapshot is older than that.

Dashboard tabs

When SLA breach follow-up is enabled for your tenant, a tab strip appears at the top of the page with two tabs:

• Overview — the KPI cards and widget grid described below. This is the existing dashboard and is unchanged.
• SLA Follow-ups — a worklist of SLA breaches that supervisors have claimed for follow-up. See SLA Follow-ups tab below.

If the feature is not enabled for your tenant, the tab strip is hidden and the dashboard shows the Overview content directly.

Page layout

The layout below describes the Overview tab.

From top to bottom:

1. SLA breach banner — appears only when there are unacknowledged breaches in the last five minutes. Dismissible per-user; reappears automatically the next time a fresh breach arrives.
2. KPI cards row — six headline metrics for today since midnight, each with a coloured stripe along the bottom edge so you can recognise them at a glance.
3. Conversation status · team capacity · channel breakdown — three-up tile row.
4. Hourly traffic · bot performance live — two-up row.
5. Agent availability — full-width table.
6. Department queue status · agent performance snapshot — two-up row.
7. Department performance · CSAT live feed — two-up row.
8. SLA breaches live — full-width table of unacknowledged breaches.

KPI cards

The six KPI cards along the top each carry a colour-coded bottom stripe:

Each card shows the value, a trend arrow with the percent change vs. yesterday, and (for cards that have one) a small breakdown line underneath. For First Response Time and Resolution Time, “lower is better” — a downward arrow there is shown as a positive trend.

Tiles

Conversation status donut

Donut chart of open conversations split across Open, Pending, Resolved, and Closed sub-states with the count on each slice.

Team capacity

Per-team open-chats and unassigned counts, with a load percentage against configured capacity.

Channel breakdown

Inbound message volume by channel for today.

Hourly traffic

Line chart of last 24 hours: messages started vs. messages closed per hour.

Bot performance (live)

Bot turn count, containment %, transfer rate, fallback rate, and resolved/transferred mix for the current hour.

Agent availability

Full-width table listing every agent with their Accepting / Busy / Away / On Break / Offline status pill and active-chat count. Counts at the top group Accepting, Away, and Offline. A search box filters by name or email. Each row has a Force Logout action — confirming it signs the agent out and returns their open chats to the queue.

Department queue status

Per-department open count, waiting count, and average wait time.

Agent performance snapshot

Top agents today by handled count, with their average FRT alongside.

Department performance

Per-department aggregate KPIs (handled count, FRT, CSAT) for today.

CSAT live feed

The most recent inbound CSAT responses across the tenant — score, agent, and timestamp.

SLA breaches live

Table of SLA breaches with conversation ID, metric (First Response · Response Time · Resolution · Queue Wait), threshold, actual time, and when it tripped. Conversation IDs are click-throughs that open the breached conversation in the inbox, where a supervisor can whisper coaching to the handling agent or transfer the chat as needed.

The per-row action depends on whether SLA breach follow-up is enabled for your tenant:

• Feature enabled — each row has a Claim button. Claiming a breach takes ownership of it for follow-up: the breach leaves this shared widget and moves to the claiming supervisor’s SLA Follow-ups tab. Admins additionally see an Assign… button, which opens a modal to assign the breach follow-up to a chosen supervisor instead of claiming it themselves.
• Feature disabled — each row has an Acknowledge button. Acknowledging clears the row from the live view with no follow-up tracking. The breach stays in the historical SLA log either way.

SLA Follow-ups tab

The SLA Follow-ups tab is a worklist of SLA breaches that have been claimed (or assigned) for follow-up. It appears in the dashboard tab strip when the feature is enabled.

Filters at the top of the tab:

• Open / Resolved sub-filter — switch between follow-ups still in progress and ones that have been completed.
• Mine / All supervisors toggle — admins only; supervisors always see just their own follow-ups.

Each row shows the breach detail (metric, threshold, actual), the handling agent, the follow-up owner, and an age indicator that turns amber after 2 hours and red after 4 hours so stale follow-ups stand out.

Each open row has two actions:

• Open chat — jumps into the breached conversation in the inbox, where the supervisor can whisper the agent.

• Mark followed up — opens a modal that requires picking a disposition before saving:

  • Coached the agent
  • Staffing gap
  • Process gap
  • No action needed
  • False alarm

  An optional note can be added alongside the disposition. Saving moves the follow-up to the Resolved filter.

A follow-up also closes automatically — moving to the Resolved filter without a disposition prompt — when the underlying conversation is resolved.

Refresh

The Refresh button in the top-right invalidates every tile at once. Individual tiles also re-poll on their own 30-second cadence; you only need Refresh when you want an immediate update.

SLA alerts

Every SLA rule under Settings → SLA Rules with Supervisor alert ticked feeds both the top-of-page breach banner (last five minutes) and the SLA Breaches Live table. When SLA breach follow-up is enabled, breaches claimed from that table flow into the SLA Follow-ups tab for tracking through to a disposition.