> ## Documentation Index
> Fetch the complete documentation index at: https://docs.digifist.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Inbox Analytics

> Track response time, conversation volume, and agent performance across your Galantis Inbox.

Inbox analytics give you visibility into how effectively your team is handling customer conversations. The metrics track both the speed and volume of your support operation — helping you identify bottlenecks, measure individual agent performance, and understand conversation patterns over time.

## What this covers

* Available Inbox metrics and what each measures
* How to interpret response time and resolution data
* Agent-level performance breakdown

## Available metrics

<Tabs>
  <Tab title="Response Time">
    **First response time**: The average time between a conversation arriving as `OPEN` and the first agent reply being sent.

    This is the primary speed metric for the Inbox. A low first response time indicates that conversations are being picked up quickly. A high first response time may indicate understaffing, routing inefficiency, or agents being overloaded with concurrent conversations.

    Monitor this metric alongside conversation volume — a rising response time during a high-volume period is expected, but a consistently high response time during normal volume suggests a structural issue.
  </Tab>

  <Tab title="Conversation Volume">
    **Inbound conversation volume**: The total number of new conversations opened during a selected time period.

    Volume data helps with staffing decisions and capacity planning. Tracking volume over time reveals patterns — peak days, seasonal spikes, or the impact of a campaign sending on inbound message rates.

    <Tip>
      Campaign sends frequently generate inbound replies. If you run a large broadcast, expect a corresponding spike in Inbox volume for the following 24–48 hours and staff accordingly.
    </Tip>
  </Tab>

  <Tab title="Agent Performance">
    **Messages sent per agent**: The total number of messages sent by each agent during a selected period.

    **Conversations resolved per agent**: The number of conversations each agent moved to `RESOLVED` status during a selected period.

    These metrics provide a per-agent breakdown of workload and output. Use them to identify high-performing agents, spot agents who may be struggling, and distribute workload more evenly through assignment routing.

    <Note>
      Messages sent and conversations resolved measure different things. An agent can send many messages while resolving few conversations — this may indicate complex or escalated cases, not low performance. Always interpret agent metrics in context.
    </Note>
  </Tab>

  <Tab title="Resolution Time">
    **Average resolution time**: The average time from a conversation opening (`OPEN`) to being marked `RESOLVED`.

    Resolution time reflects the full lifecycle of a conversation, from first contact to close. It is a more complete picture of support efficiency than response time alone — a fast first response followed by a long unresolved thread still represents a poor customer experience.

    A high average resolution time may point to: conversations being left in `PENDING` without follow-up, complex issues requiring multiple back-and-forth messages, or agents not closing resolved conversations promptly.
  </Tab>
</Tabs>

## Best practices

* **Track first response time as your primary health metric.** It is the metric most directly in your team's control and most directly experienced by the customer.
* **Compare volume to resolution time.** If volume increases and resolution time holds steady, your team is scaling well. If resolution time rises with volume, capacity may need adjustment.
* **Review agent performance in aggregate, not in isolation.** Individual message counts vary based on conversation complexity. Look for meaningful outliers rather than minor differences.
* **Account for campaign sends when reviewing volume data.** Spikes in inbound conversations shortly after a broadcast are expected — do not interpret them as anomalies.

## Related guides

* [Assignment & Routing](./assignment-routing) — How conversation distribution affects agent metrics
* [Conversation Lifecycle](./conversation-lifecycle) — How status transitions relate to resolution time measurement
* [Inbox Add-on Billing](/whatsapp/billing/add-ons/inbox) — How billable conversations are tracked alongside analytics