Average resolution time

Average resolution time (ART, sometimes called mean time to resolution or MTTR in support contexts) is the average elapsed time between a ticket being opened and being marked as resolved. It is the broadest measure of support speed, capturing everything that happens during the life of a ticket: first response wait, agent investigation, customer replies, internal escalations, and final confirmation.

This metric matters because it represents the customer's total wait for a solution. A customer who creates a ticket on Monday and receives confirmation of resolution on Thursday experienced a 3-day resolution time regardless of whether the actual agent work time was 30 minutes. The customer's perception is shaped by the full window, not by the active work time alone.

Resolution time is also a critical capacity metric. Long resolution times mean more tickets remain open simultaneously, increasing the ticket backlog, raising context-switching costs for agents, and creating a growing backlog that can become self-reinforcing. When resolution times increase by even 10%, the cumulative effect on open ticket volume can be dramatic.

However, resolution time must be interpreted carefully. A team that resolves 90% of tickets in 2 hours but has a 10% tail that takes 5 days may show an average of 14 hours. The average hides the fact that most customers have an excellent experience while a small group has a poor one. Tracking percentiles (P50, P90, P95) alongside the average provides a much more accurate picture.

For each resolved ticket, calculate the elapsed time from the moment it was created (or the customer's first message) to the moment it was marked as resolved. Sum those durations across all tickets resolved in the period, then divide by the count of resolved tickets.

Decide whether to measure in calendar time or business hours, and whether to include time spent waiting for the customer to reply. Many teams exclude "waiting on customer" status from the calculation, since the support team cannot control how quickly the customer responds. Others include it, reasoning that the customer still experiences the delay.

Resolution time is the sum of every delay and every working period in a ticket's life. A metric tree breaks it into phases and reveals where the most time is consumed.

Each branch of the tree represents a phase of the ticket lifecycle. If first response wait is the dominant contributor, the fix is staffing and routing. If investigation takes too long, the fix is better tooling and knowledge resources. If back-and-forth exchanges are the problem, the fix is better intake forms and clearer agent communication. If escalation delays dominate, the fix is escalation process improvement or expanded frontline authority.

The tree also reveals interaction effects. Improving intake quality reduces both back-and-forth exchanges and investigation time. Expanding agent authority reduces both escalation delays and first response time wait (since fewer escalated tickets means shorter queues at higher tiers). These cascading benefits are visible in the tree but invisible in a single resolution time number.

1. 1

   Improve first contact resolution rate

   Tickets resolved on the first interaction have the shortest possible resolution time. Every percentage point improvement in FCR directly reduces average resolution time by eliminating multi-day back-and-forth tickets from the denominator.

2. 2

   Capture comprehensive information at intake

   Require structured information in the ticket creation form: steps to reproduce, error messages, account identifiers, and screenshots. Each missing piece adds a round trip of clarification that can add hours or days to resolution time.

3. 3

   Reduce escalation delays with clear escalation paths

   Define explicit criteria for when and how tickets should be escalated. Set SLAs for escalation response times. Ensure escalation queues are staffed and monitored with the same rigour as frontline queues.

4. 4

   Implement SLA-driven prioritisation

   Not all tickets are equal. Assign priority levels based on impact and urgency, and ensure the queue is ordered by SLA proximity rather than arrival order. This prevents low-priority tickets from blocking high-priority ones.

5. 5

   Track and address the long tail

   Average resolution time can improve while a subset of tickets takes increasingly long. Monitor P90 and P95 resolution times alongside the average. Investigate tickets that remain open beyond a threshold and implement processes to prevent them from stalling.

KPI Tree lets you model resolution time as the sum of its component phases: first response wait, investigation, back-and-forth, and escalation. Each phase can be decomposed further by channel, priority, team, and issue category.

Connect resolution time to its upstream drivers (staffing, routing, knowledge base coverage) and downstream impacts (open ticket volume, customer satisfaction score, and cost per ticket). When resolution time increases, the tree shows which phase expanded, in which context, and who is best positioned to address it. This transforms resolution time from a lagging indicator into an actionable diagnostic framework.