Qvasa reports several time-based metrics across email, messaging, and Quiq. Many sound similar but each one measures a different thing. This guide walks through each channel's lifecycle, lists the exact metric names you'll see in your dashboards and quick stats, and helps you pick the right one.

How to read the diagrams. Each section has a lifecycle diagram with events along the top and metrics as bars underneath. Each bar shows the two events its metric measures between. Metric names in the bars match the names you'll see in your dashboard widgets and quick stats.

Metrics that apply to ticket-based channels (Zendesk Support, etc.).

The time between a ticket arriving and the first public agent comment. The customer-facing measure of how fast you answer a brand-new ticket — includes any wait before an agent was assigned.

Median variant: Median Comment First Touch Time.

Same idea, but measured from when the agent was assigned. Useful for evaluating an individual agent's responsiveness without penalizing them for queue time.

Variants: Median Comment FTT After Assignment (median), Avg Comment First Touch Time After First Assignment (counts only the first agent who picked up the ticket).

Within a single ticket, the average gap between each customer comment and the agent's reply. We compute this per ticket first, then average across tickets in your date range. Captures the back-and-forth pace of conversations, not first-response speed.

Variants: Avg Comment Agent Reply Time On Ticket Assignments (computed per agent-assignment instead of per ticket — different denominator on multi-handoff tickets), Median Median Comment Reply Time (per-ticket median, then median across tickets).

The average sum of every interval where the customer was waiting for an agent reply, totaled per ticket. Different from Avg Comment Agent Reply Time — this adds up all the waiting periods on a ticket rather than averaging them.

Total elapsed time from ticket creation to ticket being solved.

Median variant: Median Time to Solve.

Same idea, measured from agent assignment. On tickets that get reassigned, this can produce a data point per agent who held the ticket.

Median variant: Median Time from Assignment to Solve. SLA-rate variant: Email Time to Solve After Assignment SLA Rate.

Available as an agent-table column. The average time tickets currently in "open" status have been sitting open, attributed to each agent. Use this to spot agents with stale open tickets, not for historical performance.

Metrics that apply to live messaging channels — Zendesk Messaging via Sunshine Conversations. Metric names use the "Sunco" prefix in your dashboards.

The time between the conversation starting and the first agent reply. Includes any wait before an agent was assigned and engaged. The customer-facing measure of how quickly someone is reached on a new conversation.

SLA-rate variant: Messaging FTT SLA Rate (configurable threshold).

The gap between an agent being assigned and the agent's first message. Use this when evaluating an individual agent's responsiveness without penalizing them for any pre-assignment time.

Variants: Sunco Agent Chats Median First Touch Time After Assignment, Sunco First Touch Time After First Assignment Avg Duration (counts only the first agent assigned), Sunco Median First Touch Time After First Assignment, First Touch After Assignment SLA Rate (within-SLA rate of the same metric).

Point-in-time variants for chats currently still open (not finished). Open Avg Reply Time shows the average reply time among the agent's currently-open chats; Open Chat FTTAA shows the longest pending First Touch After Assignment among the agent's open chats. Use these as real-time monitoring metrics — they don't fit a historical date range.

Within a single chat session, the average gap between each customer message and the agent's reply. Aggregated across chat sessions in your date range.

The percentage of chat sessions where the time from agent assignment to the customer's last message was within your configured SLA threshold. Trailing agent messages don't extend it — if the agent says "anything else?" after the customer's last reply, that doesn't count toward Handle Time.

As a duration: Sunco Avg Assignment to Last End User Message Duration, Sunco Median Assignment to Last End User Message Duration.

A wider window than End User Handle Time — measured from agent assignment to the agent's own last message. Includes any tail-end agent messages after the customer stopped responding.

Median variant: Sunco Median Assignment to Last Agent Message Duration.

The trailing tail — gap between the customer's last message and the agent's last message. Captures how long the agent kept the chat open after the customer was done. Equal to the difference between Sunco Avg Assignment to Last Agent Message Duration and Sunco Avg Assignment to Last End User Message Duration.

Median variant: Sunco Median Last End User Message to Last Agent Message Duration.

The total length of a chat session from start to last message. Includes any pre-assignment time (bot interactions, queue) plus the entire agent-engaged portion.

Median variant: Sunco Agent Chat Median Duration.

If your account uses Zendesk's omnichannel routing, Sunco conversations also have an offer-to-accepted window — the time between when an agent is offered the conversation and when they accept it (which becomes the assignment). The diagram above shows this as a small green bar between the Offered and Agent assigned events. See the Omnichannel routing — Offer to Accepted section below for the metric details.

Metric names are prefixed with "Quiq:" in your dashboards. Quiq has its own pre-calculated handle time metrics, including a concurrency-adjusted variant.

Conversation start to the agent's first message.

Median variant: Quiq: Median Agent Time to First Response.

Within a conversation, the average gap between each customer message and the agent's reply.

Median variant: Quiq: Median Agent Response Time.

Wall-clock time from agent assignment to the last message in the conversation. Doesn't account for how busy the agent was.

Median variant: Quiq: Median Agent Handle Time Basic.

The same wall-clock duration, divided by how many other conversations the agent was handling simultaneously. An agent juggling three chats at once won't have each one's handle time inflated by the time spent on the other two. Usually the fairer metric for evaluating individual agent throughput in a high-volume messaging channel.

Variants: Quiq: Median Agent Handle Time Concurrency Adjusted, Quiq Handle Time SLA Rate (within-SLA rate of the basic handle time at a configurable threshold).

From agent assignment to the agent's first message. Quiq's analogue of First Touch Time After Assignment.

Median variant: Quiq: Median Assignment to First Message Duration.

From agent assignment to the last message of the conversation, regardless of who sent it.

Median variant: Quiq: Median Assignment to Last Message Duration. SLA-rate variant: Quiq First Touch After Assignment SLA Rate (corresponding within-SLA rate).

Quiq's analogue of Sunco End User Handle Time as a duration: agent assignment to the customer's last message. Stops before any trailing agent messages.

Median variant: Quiq Median Assignment to Last End User Message Duration.

Quiq's trailing tail — gap between the customer's last message and the agent's last message. Equal to Quiq Avg Assignment to Last Message Duration minus Quiq Avg Assignment to Last End User Message Duration.

Median variant: Quiq Median Last End User to Last Agent Message Duration.

Total length of a Quiq conversation from start to last message. The widest of the Quiq duration metrics — analogous to Sunco Agent Chat Avg Duration.

Channel-agnostic. Whenever Zendesk's omnichannel routing offers a work item to an agent — a messaging conversation, a call, an email ticket, a chat — there's a window between when the offer arrives and when the agent accepts it. These metrics measure that window. Requires omnichannel routing to be enabled on your account.

Average time between when an agent first receives an offer for a work item and when that same agent accepts it. The accepting agent is assigned to the ticket the moment the offer is accepted. Useful for measuring routing latency or agent pickup speed across any channel routed via omnichannel.

Median variant: Offer to Accepted Median Duration.

Voice channel metrics. The lifecycle is shorter than messaging — there's no back-and-forth reply pacing.

Total duration of a call from start to end.

Variants: Median Call Duration, Total Call Duration (sum across calls in your range).

Time the agent was actively on the call — from connection to end. Excludes any pre-connection ringing or routing time.

Median variant: Agent Assigned to Call Ended Median Duration.

These metrics measure agent state over time — not events on a particular conversation. They appear in the agent table and Quiq agent session tables.

Time the agent was logged in and marked available to take new conversations. Quiq agent session metric.

Time the agent was available to continue handling conversations they already had, but not to take new ones (e.g. an agent in "wrap-up" mode). Quiq agent session metric.

Time the agent was actively engaged in one or more conversations. Quiq agent session metric.

Time the agent was logged in but marked unavailable (break, lunch, etc.). Quiq agent session metric.

The agent's full session length — sum of available, in-conversation, and unavailable time.

Quiq's handle time, sliced by the agent's status when the handle time accrued. Each variant exists in both Basic (raw wall-clock) and Concurrency (divided by parallel chats) flavors:

Session-level rollups of first-response and per-message response times across all of an agent's conversations in the session. Distinct from the conversation-level Quiq: Average Agent Time to First Response and Quiq: Avg Agent Chat Response Time — same concept, different aggregation grain.

Time the agent spent in a particular work-item status or unified status (e.g. "Online", "Busy", "Away"). Choose a status when configuring the column. Useful for capacity reporting and shift analysis.

In the Quiq Status Records table, the time spent in each individual status row. Use this for the granular view of an agent's status changes throughout a session.

"Handle time" is the most overloaded label across channels:

If two dashboards report different "handle time" values, this is usually why.

Most metrics ship in several flavors. Knowing which variant a dashboard is using helps explain differences between two seemingly identical numbers.

Averages are pulled by long-tail outliers — one four-hour reply will lift the average noticeably, even if 99% of replies were quick. Medians describe the typical experience and are less sensitive to outliers. We expose both for most metrics. Use median when you care about typical performance, average when you care about totals or worst-case impact.

Calendar-hours metrics count all elapsed time including nights, weekends, and holidays. Business-hours variants exclude time outside your configured business hours. Use business hours for evaluating agent or team performance; calendar hours for what the customer actually experienced.

Some metrics report the duration itself ("avg first response was 6 minutes"). Others report the rate of meeting a target ("82% of tickets met the 10-minute SLA"). Both are useful — durations tell you the typical magnitude, rates tell you how often you're hitting your target. SLA-rate metrics in this article include Messaging FTT SLA Rate, Sunco End User Handle Time SLA Rate, and Quiq Handle Time SLA Rate.

When a ticket is reassigned between agents, some metric variants only count the first agent who picked it up. Others evaluate every agent who held the ticket. Examples: Avg Comment First Touch Time After First Assignment only counts the original handler; Avg Comment First Touch Time After Assignment evaluates every assignment.

For a ticket that two agents worked on, you can ask "what was the reply time for this ticket?" (one number) or "what was the reply time for each agent on this ticket?" (two numbers). Examples: Avg Comment Agent Reply Time uses the per-ticket grain; Avg Comment Agent Reply Time On Ticket Assignments uses the per-agent-on-ticket grain.

A quick reference for picking the metric that matches the question you're trying to answer.

If two dashboards show different values for what looks like the same metric, the difference is almost always one of these:

If you're still stuck, reach out to your Qvasa support contact and we'll help you reconcile.