The quick download
MTTR, MTBF, MTTD, and MTTF each measure a different stage of reliability, from detection to recovery to long-term system performance.
-
MTTD = how fast you detect, MTTR = how fast you fix, MTBF = how often systems fail, and MTTF = how long components last.
-
Inconsistent MTTR definitions or incorrect MTBF calculations lead to misleading insights.
-
Measure a baseline for at least one full quarter before you change tools or processes, then re-measure with the same definitions so you can show what actually improved.
-
Recommendation: Write down one definition per metric, including what counts as a failure and when each clock starts and stops. Then report the metrics in the dashboards and incident reviews your team already uses, so a rising MTTR or falling MTBF points to a specific fix.
MTTR, MTBF, MTTD, and MTTF all measure system reliability, but each one answers a different question:
- MTTD measures how quickly you detect a problem after it occurs.
- MTTR shows how fast you fix it.
- MTBF measures the average operating time between failures for a repairable asset.
- MTTF estimates the average time a non-repairable component operates before it fails.
Together, these metrics cover the full path from detection to recovery to long-term reliability. People often mix them up because the names are so similar.
In this guide, we’ll break down each metric, explain when to use it, and show how they fit together in real-world monitoring and incident response.
Definitions of Reliability Metrics
MTBF measures how often a repairable system fails, while MTTR measures how long it takes to repair it after each failure. Together, they show how frequently failures happen and how long each one affects your environment.
Here we see both with eight related metrics that cover detection, acknowledgment, diagnosis, service restoration, and verification.
What is MTTR?
MTTR stands for mean time to repair. It measures the average time it takes to repair a system and return it to working condition after a failure. The same acronym is also used for:
- mean time to recovery
- mean time to resolve
- mean time to respond, so state which one you mean before you report on it
MTTR is important because it tells you how long each failure keeps a system degraded or down. Combined with MTBF, it determines availability, and its trend shows whether your runbooks, tooling, and on-call process are actually shortening outages.
The four MTTR variants cover different spans of an incident:
- Mean time to repair: time spent repairing the failed system, including testing, until it works again
- Mean time to recovery or restore: time from the moment the failure occurs until service is fully restored for users, including detection and diagnosis
- Mean time to resolve: time from incident start until the incident is fully resolved, including follow-up work to prevent recurrence
- Mean time to respond: time from the first alert until the system is recovered, excluding any lag in the alerting system. It differs from MTTA, which stops as soon as someone acknowledges the alert.
MTTR formula: MTTR = total repair time / number of repairs
Let’s say three failed drives need to be swapped out. Two take 5 minutes each to replace, and one takes 6 minutes because the drive sled is stuck. Add the repair times and divide by three:
(5 + 5 + 6) / 3 = 5.3 minutes
Your average repair time is 5.3 minutes. Tracked over time, this shows whether a repeatable fix is getting faster or slower.
Important: The right MTTR for your team depends on how you handle incidents and what you want to measure.
- Infrastructure and operations teams usually use mean time to repair or restore, because their job is fixing systems and getting services back quickly.
- Service and support teams often use mean time to resolve, since they own the issue from first report to full resolution, including investigation and customer communication.
- Incident response and on-call teams often track mean time to respond to measure how quickly they move from the first alert to recovery.
What is MTBF?
MTBF stands for mean time between failures. It measures the average operating time between failures for a repairable asset. The higher the MTBF, the less often the system fails.
When failures occur at a steady rate, MTBF is the inverse of the failure rate. A server with an MTBF of 180 hours has a failure rate of 1/180, or about 0.0056 failures per hour.
MTBF formula: MTBF = total uptime / number of failures
Imagine a production server runs for 720 hours over a month and fails 4 times during that period. Divide the total uptime by the number of failures:
720 / 4 = 180 hours
The server runs for about 180 hours, on average, before the next failure. Count only operating time. If repair time slips into the total, MTBF looks better than the system really is.
What is MTTF?
MTTF stands for mean time to failure. It measures the average time a non-repairable component lasts before it fails. IT teams use it for non-repairable systems and parts that get replaced rather than repaired, such as hard drives, batteries, or sensors, to estimate lifespan and plan refresh cycles.
MTTF formula: MTTF = total operating time / number of failures
For example, your team replaces three failed drives in a storage array. One lasted 2.1 years, one lasted 2.7 years, and one lasted 2.3 years. To find the average lifespan, add those numbers and divide by three:
(2.1 + 2.7 + 2.3) / 3 = 2.37 years
The drives lasted about 2.37 years on average before failing, which gives you a practical baseline for replacement planning.
What is MTRS?
MTRS stands for mean time to restore service. It measures the average time from a service failure until the service is fully working again for users. It covers the whole outage, including detection, diagnosis, repair, and verification, instead of repair work alone.
MTRS formula: MTRS = total unplanned service downtime / number of service failures
Suppose a customer-facing application has four outages in one quarter. The outages last 3 hours, 2 hours, 4 hours, and 1 hour. Add the downtime and divide by four:
(3 + 2 + 4 + 1) / 4 = 2.5 hours
It takes 2.5 hours, on average, to restore service after an outage. MTRS excludes planned maintenance. If you need a number that includes scheduled downtime, use MDT.
What is MTBSI?
MTBSI stands for mean time between service incidents. It measures the average time from the start of one service incident to the start of the next. Because that span includes the outage itself, MTBSI equals MTBF plus MTRS.
MTBSI formula: MTBSI = MTBF + MTRS
If a database server has an MTBF of 300 hours and an MTRS of 4 hours, the calculation looks like this:
300 + 4 = 304 hours
A new service incident starts about every 304 hours, measured from the start of one incident to the start of the next.
What is MTTD?
MTTD stands for mean time to detect. It measures the average time between when a failure starts and when your monitoring raises an alert, or someone notices the problem.
MTTD formula: MTTD = total time from failure to detection / number of failures
Your team reviews five incidents. The time between failure and detection was 4 minutes, 6 minutes, 3 minutes, 5 minutes, and 7 minutes. Add those times and divide by five:
(4 + 6 + 3 + 5 + 7) / 5 = 5 minutes
Your average detection time is 5 minutes. A high MTTD usually points to gaps in monitoring coverage, check intervals that are too long, or thresholds that only fire after users are already affected.
Track MTTD per service tier for at least a quarter before you tune alerting. That baseline tells you whether a change actually shortened detection or just moved the number around.
What is MTTI?
MTTI stands for mean time to identify. It measures the average time from detection to identifying which component or service is failing. It answers “where is the problem?”
For example, the database is failing, not the load balancer, before anyone knows why.
MTTI formula: MTTI = total time from detection to identifying the failing component / number of incidents
Assume your team handles four performance incidents in a month. It takes 35 minutes to identify the failing component in the first, 20 minutes in the second, 10 minutes in the third, and 15 minutes in the fourth:
(35 + 20 + 10 + 15) / 4 = 20 minutes
Your team needs 20 minutes, on average, to pinpoint where the problem sits.
What is MTTK?
MTTK stands for mean time to know. It measures the average time from detection to understanding the root cause, meaning why the component failed. MTTI tells you the database is failing, or it tells you a schema change locked a table.
MTTK formula: MTTK = total time from detection to root cause identification / number of incidents
Say your team investigates three incidents. Root cause analysis takes 1.5 hours for the first, 1.75 hours for the second, and 1 hour for the third:
(1.5 + 1.75 + 1) / 3 = 1.42 hours
It takes about 1.42 hours, on average, to move from detection to root cause. MTTK is most useful when alerts fire quickly, but troubleshooting still drags on.
What is MDT?
MDT stands for mean downtime. It measures the average length of every period a system is non-operational, both planned and unplanned. Unlike MTRS, it includes scheduled maintenance windows, so it reflects total service impact.
MDT formula: MDT = total downtime (planned and unplanned) / number of downtime events
Imagine a critical internal application goes down four times in a month. The outages last 2 hours, 30 minutes, 1 hour, and 25 minutes. Convert everything to minutes, then divide by four:
(120 + 30 + 60 + 25) / 4 = 58.75 minutes
The application is unavailable for about 59 minutes, on average, each time it goes down.
What is MTTA?
MTTA stands for mean time to acknowledge. It measures the average time from when an alert fires to when someone acknowledges it. In incident management, MTTA reflects how quickly on-call staff pick up new issues.
MTTA formula: MTTA = total time to acknowledge alerts / number of incidents
Say your on-call team receives four alerts. They acknowledge them in 2 minutes, 4 minutes, 3 minutes, and 1 minute:
(2 + 4 + 3 + 1) / 4 = 2.5 minutes
Your team acknowledges alerts in 2.5 minutes on average. When MTTA rises, check alert routing and alert volume before you look at individual responders.
What is MTTV?
MTTV stands for mean time to verify. It measures the average time it takes to confirm that a fix actually worked and the service is stable. It tracks the final step of incident handling.
MTTV formula: MTTV = total time to verify fixes / number of resolved incidents
Suppose your team resolves three incidents, then spends 8 minutes, 12 minutes, and 10 minutes verifying that services are healthy again:
(8 + 12 + 10) / 3 = 10 minutes
Verification takes 10 minutes on average. If this number is high, add automated health checks or synthetic tests that confirm recovery from the user’s side.
Metric Comparisons
Each metric answers one question about an incident. Choose the metric that matches the decision you need to make.
| The question you need to answer | Use | Don’t use it when |
|---|---|---|
| How often does this system fail? | MTBF | The part is replaced, not repaired. Use MTTF. |
| How long do replaceable parts last? | MTTF | The system is repaired and returned to service. Use MTBF. |
| How long does each failure take to fix? | MTTR (say which R you mean) | You need to know how fast you noticed the failure. Use MTTD. |
| How fast do we notice a failure? | MTTD | The delay is a person picking up the alert. Use MTTA. |
| How fast does someone pick up an alert? | MTTA | You want to measure the repair itself. Use MTTR. |
| How much of the time is the service usable? | Availability = MTBF / (MTBF + MTTR) | You have only one or two incidents in the period. The number won’t be meaningful yet. |
In an incident, the clocks run in this order: failure, detection (MTTD), acknowledgment (MTTA), identification (MTTI), repair (MTTR), verification (MTTV).
MTTR vs MTBF
MTBF measures how often a repairable system fails, and MTTR measures how long it takes to repair it each time. In MTBF vs MTTR terms, MTBF measures reliability and MTTR measures maintainability.
You want MTBF high (fewer failures) and MTTR low (shorter outages). Used together, they give you availability.
| Metric | What it measures | Formula | Ideal direction | Best for | Common mistake |
|---|---|---|---|---|---|
| MTTR (Mean Time to Repair) | Average time to repair a system after a failure | Total repair time / number of repairs | Lower is better | Measuring repair and recovery efficiency | Mixing different definitions (repair vs resolve vs respond) |
| MTBF (Mean Time Between Failures) | Average time a system runs before failing | Total uptime / number of failures | Higher is better | Measuring system reliability and stability | Including repair time in the calculation |
How MTTR and MTBF work together
MTTR and MTBF work together to show how often systems fail and how quickly teams recover.
| Scenario | What it means | What to fix |
|---|---|---|
| Low MTBF + High MTTR | Systems often fail and take a long time to recover | Improve reliability and speed up recovery processes |
| Low MTBF + Low MTTR | Systems fail often but are fixed quickly | Reduce failure frequency and address the underlying instability |
| High MTBF + High MTTR | Systems fail rarely but recovery is slow | Improve response and recovery processes |
| High MTBF + Low MTTR | Systems are stable and recover quickly | Ideal state to maintain |
This combined view helps teams decide whether to focus on preventing failures, shortening recovery, or both.
Calculating MTTR and MTBF
The example below uses one server and one month of data to calculate MTBF, MTTR, MTTD, MTTA, and availability together.
Suppose a team tracks one production server over a month. The server is up for 720 hours and fails 4 times. Repairs take 8 hours in total. For each failure, monitoring took 4, 6, 3, and 7 minutes to detect it, and the on-call engineer took 2, 4, 3, and 1 minutes to acknowledge the alert.
- MTBF = total uptime / number of failures = 720 / 4 = 180 hours
- MTTR = total repair time / number of repairs = 8 / 4 = 2 hours
- MTTD = (4 + 6 + 3 + 7) / 4 = 5 minutes
- MTTA = (2 + 4 + 3 + 1) / 4 = 2.5 minutes
- Availability = MTBF / (MTBF + MTTR) = 180 / (180 + 2) = 98.9%
This server fails about every 180 hours, and each failure takes about 2 hours to repair.
There are two ways to raise its availability.
- Cutting MTTR to 1 hour gives 180 / 181 = 99.4%.
- Doubling MTBF to 360 hours gives 360 / 362 = 99.4%.
The result is the same, so choose the lever that’s cheaper for your team: faster recovery through runbooks and automation, or fewer failures through maintenance and redundancy. For more on how availability differs from raw uptime, see uptime vs. availability.
Convert all times to the same unit before you calculate. If your team tracks mean time to recovery rather than repair, add the detection and acknowledgment time to each outage.
How to Improve Each Metric in Practice
You improve MTBF by preventing failures and MTTR by shortening each recovery.
To improve MTBF (reduce failures):
- Use preventive and predictive maintenance to catch issues early
- Replace unreliable components with higher-quality hardware or services
- Add redundancy and failover so a single component failure doesn’t take the service down, which also raises availability
- Monitor performance trends to spot recurring issues before they cause outages
- Improve testing and deployment processes to reduce production errors
To improve MTTR (recover faster):
- Set up real-time monitoring and alerting so repair starts sooner
- Use runbooks and incident response playbooks to guide faster fixes
- Automate recovery for known, repeatable failures. Start with rule-based actions such as restarting a hung service, clearing a full log volume, or failing over to a standby node, and require approval and a rollback step for anything broader. In LogicMonitor, simple actions can run directly from the collector, broader actions can run through tools you already use like Ansible, and Edwin AI can recommend and coordinate next steps within approvals, audit trails, and validation. See how this works in Reduce MTTR.
- Keep spare parts on site for critical hardware, such as drives, power supplies, and optics, so repairs don’t wait on shipping
- Improve logging and observability to speed up root cause analysis
- Train teams with incident simulations to reduce response delays
Bechtle Suisse, an IT services provider, moved from SNMP-only monitoring to LogicMonitor and reported quicker, more accurate root cause analysis because of the wider range of metrics collected per device. For AI-specific tactics, see how to reduce MTTR with AI.
Setting Targets for MTTR and MTBF
There’s no single good MTTR or MTBF. A useful target depends on how critical the service is, how severe the incident is, and whether the asset is repaired or replaced.
Set targets from your own data:
- Measure a baseline: Track at least one full quarter for each service tier and severity level, using the same failure definitions every time.
- Set tighter targets for critical services: A customer-facing checkout service needs a shorter MTTR than an internal reporting tool.
- Define three bands: Agree on what counts as good, acceptable, and needs attention for each tier, and alert on the last one.
- Judge the trend, not one month: Compare rolling quarters, and compare before and after each process or tooling change.
Treat vendor MTBF ratings as a starting point. Vendors measure them under test conditions, so compare them with the MTBF you see in your own environment before you plan maintenance or replacements around them.
MTTF vs. MTBF
MTTF applies to things you replace, and MTBF applies to things you repair. “To failure” means the item’s life ends at the failure, while “between failures” means it returns to service and can fail again.
Use the remedy to decide which one to track. If a failed drive goes in the bin, track MTTF for the drive. If a switch is brought back by swapping its power supply, track MTBF for the switch and MTTF for the power supply. One system often needs both.
Vendors publish MTTF and MTBF figures from test conditions, and you can’t change a drive’s MTTF. You can run drives in RAID so a single drive failure doesn’t become an outage, which raises the MTBF of the service and helps you drive down MTTR for issues within your infrastructure.
MTTD vs. MTTI
MTTD and MTTI are back-to-back clocks. MTTD runs from the failure until the alert fires. MTTI starts when the alert fires and stops when the team identifies which component is failing.
For example, a database fails at 2:00 p.m., the alert fires at 2:05 p.m. (MTTD of 5 minutes), and the team confirms the database is the failing component at 2:25 p.m. (MTTI of 20 minutes). A short MTTD with a long MTTI means alerting works, but alerts arrive without enough context, such as topology or recent changes, to point at the cause.
MTTD vs. MTTA
MTTD measures how long it takes your monitoring to detect a failure, and MTTA measures how long it takes a person to acknowledge the resulting alert. Tooling mostly sets MTTD: coverage, check frequency, and thresholds. People and processes mostly set MTTA: on-call routing, escalation rules, and alert volume.
In LogicMonitor, MTTD is the average time from when a failure happens to when the LogicMonitor platform raises an alert. MTTA is the additional time until an on-call engineer acknowledges it.
If MTTA increases while MTTD stays flat, look for alert fatigue. When responders receive many low-value alerts, they learn to wait before acting. Grouping duplicate alerts and routing them by service owner usually lowers MTTA faster than adding more monitoring, and it gives your incident management process a cleaner starting point.
MTTF (Failure) vs. MTTR: Mean Time to Failure vs. Mean Time to Repair
MTTF measures how long a component runs before it fails, and MTTR measures how long it takes to repair the system after that failure. They measure different things, so they’re most useful side by side when you plan spares and availability.
Take the drives from earlier. They last 2.37 years on average before failing (MTTF), and each swap takes 5.3 minutes (MTTR). You raise MTTF by buying better drives or controlling heat and vibration. You lower MTTR by keeping spare drives on hand and following a tested replacement runbook.
MTTF (Fix) vs. MTTR: Mean Time to Fix vs. Mean Time to Repair
Mean time to fix and mean time to repair mean the same thing, and mean time to repair is the more common term. If you do use mean time to fix, spell it out in reports, because the acronym MTTF usually means mean time to failure.
MTRS vs. MTTR: Mean Time to Restore Service vs. Mean Time to Repair
MTRS measures the full outage, from failure until service is restored for users, while MTTR (repair) measures only the repair work. MTRS is always equal to or longer than MTTR because it also includes detection, diagnosis, and verification.
Go back to the three failed drives. Repairs took 5, 5, and 6 minutes, so MTTR is 5.3 minutes. Now suppose each failed drive took 3 minutes to find and diagnose before the swap began. MTRS adds that time:
((5 + 5 + 6) + (3 + 3 + 3)) / 3 = 8.3 minutes MTRS
Users experience MTRS, not MTTR. If the failed drives had gone unnoticed for days while slowing the array, MTTR would still read 5.3 minutes. Report MTRS when you discuss service health and user impact, and MTTR when you want to improve repair efficiency.
Common Pitfalls When Using MTTR and MTBF
MTTR and MTBF are only useful if you measure them consistently. These are the mistakes that most often produce misleading numbers:
- Inconsistent MTTR definitions: Teams often mix repair, recovery, resolution, and response under one metric, which makes comparisons unreliable. Choose one definition and use it consistently.
- Repair time inside MTBF calculations: MTBF should only measure uptime between failures. Downtime in the total inflates the metric and overstates reliability.
- Averages without context: A single long outage can skew MTTR, while frequent small failures may not stand out in MTBF. Review distributions and trends alongside the average.
- Same average, different pattern: Two services can share a 2-hour MTTR while one has many 10-minute blips and the other a rare 10-hour outage. Report the median and 90th percentile alongside the mean.
- One-sided optimization: A lower MTTR with no change in MTBF still leaves users with frequent disruptions. A higher MTBF with no change in MTTR still means long outages when failures do occur.
- A better number, not a better system: Faster ticket closure or reclassified incidents lower MTTR without fixing anything. Track repeat incidents next to MTTR to confirm the improvement is real.
- Hidden detection and response delays: MTTR often depends on how quickly issues are detected and acknowledged. A high MTTD or MTTA increases total downtime even when the repair itself is fast.
- Numbers with different scopes: One team counts partial outages, another only full failures. Agree on the definitions in “Standardize your metrics” before you compare teams or quarters.
- Metrics without follow-up actions: Metrics alone don’t improve performance. Tie each MTTR and MTBF review to a concrete action, such as a monitoring gap to close, a runbook to automate, or a maintenance schedule to change.
The Role of CMMS and EAM Systems in Managing Reliability Metrics
Computerized Maintenance Management Systems (CMMS) and Enterprise Asset Management (EAM) software track reliability and failure metrics for physical assets by tying each failure to a work order and an asset history. They are the usual home for maintenance KPIs in facilities and manufacturing. Common features include:
- Maintenance scheduling: Automate preventive maintenance tasks to reduce unexpected breakdowns
- Asset performance monitoring: Track asset condition in real time to detect issues early
- Data analysis and reporting: Use historical data to spot trends and predict future performance
In IT operations, monitoring, alerting, ITSM, and AIOps tools do the same job. Monitoring records when a failure starts and when it was detected (MTTD), the alerting or on-call tool records acknowledgment (MTTA), and the ITSM ticket records when repair work started and ended (MTTR). Connected, these tools give your incident management process accurate timestamps without anyone running a stopwatch.
These metrics become actionable when they are part of daily operations. You can track MTTR and MTBF on dashboards, set thresholds that trigger alerts, and use the metrics to prioritize incidents by impact.
CMMS and EAM systems connect these metrics directly to execution. When a failure occurs, they generate work orders, link the incident to the asset’s history, and flag recurring issues.
Over time, this data feeds trend analysis and root cause reviews, turning reliability metrics into inputs for maintenance planning and continuous improvement.
Where MTTR and MTBF Are Most Useful
How you use MTTR and MTBF depends on how failures affect users, systems, or production. Common use cases include:
- SaaS and cloud operations: MTTR tracks how quickly incidents are resolved during outages, while MTBF shows how stable services are between deployments and infrastructure changes.
- DevOps and CI/CD pipelines: MTBF shows how often deployments introduce failures into production, while MTTR measures how quickly you can roll back or stabilize systems after a bad release.
- Infrastructure and IT operations: MTBF highlights how often systems fail, while MTTR reflects how efficiently teams restore services during incidents.
- Manufacturing and maintenance: MTBF tracks equipment reliability and guides preventive maintenance schedules, while MTTR shows how long production lines stay down when a machine fails.
- Field service and support: MTTR measures how quickly customer issues are resolved, while MTBF identifies recurring failures in products or equipment.
Tracked together, the two metrics show where to invest: fewer failures, faster recovery, or both.
Put your reliability metrics to work
Use MTTR, MTBF, and MTTD to spot where incidents are taking too long, happening too often, or going undetected. Then turn those trends into targeted improvements across monitoring, alerting, and response.
LogicMonitor can help your team connect reliability insights to faster detection and recovery.
Track MTTR, MTBF, and related metrics in one place
Get real-time visibility, faster detection, and governed automation that turn these metrics into action.
FAQs
1. How can I decide whether to use MTBF or MTTF when analyzing my system?
If your system or component is repairable, use MTBF to measure the time between failures. If it’s non-repairable and gets replaced after failing (like a lightbulb or hard drive), use MTTF instead.
2. When should MTTD take priority over MTTR in monitoring systems?
MTTD should take priority when it is the primary driver of the downtime. If systems fail silently or alerts are slow, reducing detection time will have a bigger impact than improving repair speed. Faster detection helps respond earlier and prevents issues from escalating.
3. What’s the difference between MTTA and MTTD in real-world response teams?
MTTD measures how long it takes a system to detect a failure, while MTTA measures how long it takes a human to acknowledge and start working on it. MTTD is usually automated, and MTTA reflects team responsiveness. Both are important because fast detection without quick action still leads to delays.
4. Can I use both MTTF and MTBF when analyzing a single system?
Yes, but they apply to different components. Use MTBF for repairable systems that can be fixed after failure, and use MTTF for non-repairable components that are replaced. In systems, both metrics are often used together to understand overall reliability.
5. How do CMMS and EAM solutions help improve reliability metrics like MTTR and MTBF?
CMMS and EAM systems improve MTTR and MTBF by connecting metrics to operations. They track asset performance, generate work orders, store maintenance history, and support preventive maintenance. This helps fix issues faster, reduce repeat failures, and improve overall system reliability.
6. Is MTTR a KPI?
Yes. MTTR is a key performance indicator (KPI) for incident management because it measures how long failures keep systems down. It is most useful as a trend tracked per service tier with one consistent definition, and paired with MTBF so faster repairs aren’t hiding more frequent failures.
7. What is a good MTTR?
There’s no universal good MTTR. The right target depends on how critical the service is, how severe the incident is, and whether you measure repair or full restoration. Measure a baseline for at least one quarter, then set tighter targets for customer-facing services. See “Setting targets for MTTR and MTBF” above.
8. How do MTBF and MTTR relate to system availability?
Availability = MTBF / (MTBF + MTTR). A server with an MTBF of 180 hours and an MTTR of 2 hours is available 180 / 182 = 98.9% of the time. You can raise availability by cutting MTTR or by raising MTBF. In this example, halving MTTR to 1 hour and doubling MTBF to 360 hours both lift availability to 99.4%.
9. What is the difference between MTTA and MTTR?
MTTA measures the time from when an alert fires to when someone acknowledges it. MTTR measures how long it takes to repair the system afterward. MTTA covers the first few minutes of a response, and MTTR covers the fix.




