MTTR and MTTA: definition and a practical example
Short answer: MTTR and MTTA are per-incident time averages: MTTA measures how long the team takes to acknowledge the alert, and MTTR how long it takes to restore the service.
In Portuguese: MTTR e MTTA
What are MTTR and MTTA?
MTTA (Mean Time to Acknowledge) is the average time between an incident’s alert and the moment someone acknowledges it and takes ownership. MTTR is the average time until the service is working again. The acronym MTTR has variations (Mean Time to Repair, to Restore, to Recover, to Resolve), so every company needs to state where the clock starts and where it stops.
The two numbers answer different questions. MTTA shows whether the alert reaches anyone: a trunk that goes down at 2 a.m. and is only acknowledged at 7 a.m. has an alerting problem, not a repair problem. MTTR shows how long the end customer goes without service, adding up reaction, diagnosis, contact with the carrier and the fix.
In contracts, they show up as response and repair times alongside the availability SLA. A common source of confusion: MTTR can be measured from the actual start of the failure or from detection. If detection depends on a customer complaining, the official MTTR looks short while the real outage was much longer.
In practice
Four incidents in one month, counted from detection:
| Incident | Until acknowledged | Until service restored |
|---|---|---|
| A, SIP trunk, 8:00 a.m. | 6 min | 40 min |
| B, IVR, 10:30 a.m. | 2 min | 15 min |
| C, 0800, 2:10 a.m. | 25 min | 95 min |
| D, DID extension, 4:45 p.m. | 3 min | 10 min |
MTTA = (6 + 2 + 25 + 3) ÷ 4 = 9 minutes. MTTR = (40 + 15 + 95 + 10) ÷ 4 = 40 minutes. Incident C, in the middle of the night, pulls both averages up. Without it, MTTA would drop to about 4 minutes, a sign that the problem lies in after-hours alerting.
How Monitor uses this
The report shows the average time to first acknowledgment of each incident, which is MTTA, and the downtime of each incident. That way you can see whether the delay is in someone noticing the alert or in fixing the failure. See alerts and escalation.
Frequently asked questions
What is the difference between MTTR and MTTA?
MTTA ends when someone takes ownership of the incident, that is, acknowledges the alert. MTTR ends when the service is working again. MTTA measures how the team and the alerting process react; MTTR measures the reaction plus diagnosis and repair. A high MTTA almost always points to an alert nobody saw, sent to the wrong person or outside business hours.
How do you reduce MTTA?
Send the alert to someone who can act, through a channel that person sees outside business hours, and escalate to the next person on the list when nobody acknowledges within a few minutes. Cutting false alarms helps too, because a team that gets many alerts with no real problem starts ignoring them. Measuring MTTA per incident shows where the process gets stuck.