The Home Dashboard shows how your support is running, from response times and customer satisfaction to agent performance and AI automation. Administrators and managers see it, and both see the same layout. Only administrators can change it.
What's on the dashboard
The default layout includes these widgets:
- KPI stats: First response time, Resolution time, Messages per ticket, Escalation rate and One touch rate.
- AI automation chart: conversations the AI agent resolved compared with conversations handled manually.
- CSAT distribution: a donut chart of your customer satisfaction ratings.
- First response time by channel: response time split by inbox or channel.
- CSAT over time: a line chart of your satisfaction score day by day.
- Support volume: Total conversations, Backlog, Closed and Unique conversations messaged.
- Leaderboard: top performers for CSAT, response time, one touch rate and resolution time.
- Agent performance table: every agent's metrics in one table.
If a widget described below isn't on your dashboard, you can add it.
Customizing your Home Dashboard →
Date range
All metrics follow the date picker in the top right, except the team workload widgets, which always show the current moment. Where it applies, the dashboard compares the selected period with the previous period of the same length.
KPI metrics
Each KPI card shows one number for the selected period.
- First response time: the median time from when a conversation was created to the first reply from a team member. If the AI agent handled the conversation first, the clock starts at the handover, so time the AI agent spent isn't counted against your team. Lower is better.
- Resolution time: the median time from when a conversation was created to when it was resolved. Conversations count in the period they were resolved, not the period they were created. Lower is better.
- Messages per ticket: outgoing messages sent in the period divided by conversations created in the period. Lower is usually better. A very high number can point to unclear replies, missing information from customers or topics that need better help articles.
- Escalation rate: the share of conversations handled by the AI agent that were handed over to your team. Reassignments between human agents aren't counted. A high rate usually means guidance or knowledge is missing for those topics.
- One touch rate: the share of conversations resolved with a single agent reply. Higher is better.
Spam is left out of First response time and Resolution time. If you turn on the business hours option, both count only time within your configured business hours.
Support volume
- Total conversations: conversations created across every inbox in the period. Use it to spot busy periods, seasonal spikes or sudden increases that may need extra staffing.
- Backlog: conversations created in the period that aren't resolved yet. That covers open conversations, conversations in your Pending list and conversations the AI agent is still handling. A growing backlog means conversations are coming in faster than your team resolves them.
- Closed: conversations resolved in the period. Compare it with Total conversations to see whether your team is keeping pace.
- Unique conversations messaged: conversations where at least one agent sent a reply in the period. A conversation counts here as soon as an agent replies, even if it isn't resolved yet.
To see everything your team holds right now, whatever the date, see Team workload below.
CSAT
The CSAT widgets show ratings from inboxes where CSAT surveys are turned on. After a conversation is resolved, the customer can rate it from 1 to 5: 5 is Excellent, 4 Good, 3 Average, 2 Fair and 1 Poor.
The CSAT score is the share of responses rated 4 or 5. If 80 out of 100 customers rated 4 or 5, your score is 80%. Only conversations with a survey response are counted.
- CSAT distribution: a donut chart of every rating received in the period, from very unhappy to very happy.
- CSAT over time: a line chart of your daily CSAT score. Use it to spot trends and drops after events such as a product issue or a busy period.
The CSAT report covers turning surveys on, response rate and scores by agent.
AI automation
- AI automated vs manual: a bar chart that splits every conversation in the period into Automated and Manual. Automated conversations were resolved by the AI agent without a human agent replying. Manual conversations needed a human agent to step in. The chart shows how the split changes over time.
- AI automation rate: the share of total volume the AI agent resolves on its own. If the AI agent resolved 300 out of 1,000 conversations, the rate is 30%. A higher rate leaves your team more time for complex conversations.
To raise your automation rate:
- Add guidance rules so the AI agent can handle more scenarios without escalating.
- Add response examples, question-and-answer pairs that teach the AI agent how to handle specific topics.
- Expand your Customer Portal. The AI agent uses your help articles as a knowledge base, so more complete articles give better answers.
- Review knowledge gaps under AI Agent → Knowledge gaps to see what customers ask about that the AI agent can't answer yet.
The AI agent report has success, handover and pending rates, with a breakdown by guidance.
Team performance
The leaderboard shows your top three agents in four categories as podium cards:
- Top CSAT: the highest customer satisfaction score in the period. Agents need at least 5 survey responses in the period to appear.
- Fastest response: the quickest median first response time. It's measured from when the conversation was created, or from the handover if the AI agent handled it first, to the first human reply. It isn't measured from assignment.
- Best one touch rate: the agents who resolve the most conversations with a single reply.
- Fastest resolution: the shortest time from open to closed.
For Fastest response, Best one touch rate and Fastest resolution, agents need at least 5 resolved tickets in the period.
The Agent performance table has one row per agent with response time, resolution time, messages per ticket, CSAT score, one touch rate and escalation rate. Use it to compare the whole team beyond the top three, and compare date ranges to track each agent's progress.
Volume and ticket mix affect these numbers. An agent who handles complex escalations can have a lower CSAT score than one who handles simple requests.
Team workload
The team workload widgets show your queue as it is right now. The date range doesn't apply to them. Unassigned conversations and conversations the AI agent is handling aren't counted.
- Total assigned: conversations assigned to an agent across the whole team. It equals Total open plus Total pending. A high number relative to your team size can mean agents are holding too many conversations at once.
- Total open: open conversations assigned to a human agent. This number can be lower than the Open count in your inbox.
- Total pending: conversations in your Pending list that are assigned to a teammate. They wait on the customer's reply, so they don't need action now, but keep an eye on them in case they go stale.
- Average per agent: conversations assigned to human agents divided by the number of users on your account, whether or not they are online. Use it to spot when the team is overloaded and needs more hands or reassignment.
The Workload by agent tile sits under Team workload and assignment and has one row per agent:
- Assigned: open and pending conversations assigned to that agent right now.
- Open: how many of those are open.
- Pending: how many of those are in the Pending list.
- Workload: how the agent's assigned count compares with the team average, shown as Idle, Light, Moderate or Heavy.
Use this tile to see how many conversations each agent has right now. There is no report of how many tickets were assigned to an agent on a past day or week. The Agents report covers a period and has no current count.
To open one agent's queue as a list, create a view with Assignee name Equal to that agent and Status Equal to Open.