Observability · Two-minute check
Reactive or proactive? Where does your operation stand?
What the research says · What moves you up one step · MTTx done right · The posture check · Sources
Every IT team wants to find problems before users do. Most still find out from a phone call. Take the two-minute check, see where you sit on the ladder, and see the next step up.
The figures and sources could not be loaded. Please reload the page.
What the research says
of leaders said they still learn about service interruptions through customer complaints, incident tickets or manual checks.
of technology executives said customers are often or always the first to detect downtime (Global 2000 companies).
of engineers’ time goes on fighting fires or addressing disruptions instead of building new features.
median time to detect and to resolve a high-impact outage. 29% said detection took an hour or more.
of organizations that had a major outage believe it could have been prevented with better management or processes.
Several of these figures come from surveys run or paid for by monitoring vendors, and most respondents are large enterprises. We show them because they are the best published numbers, and we label them so you can weigh them.
What moves you up one step
Reactive → Proactive
- Make the tools find incidents first, starting with the services users complain about.
- Give every alert a named owner.
- Cut noise until every page is worth acting on.
- Start measuring time to detect and time to restore.
Proactive → Predictive
- Alert on how fast an error budget is being used, not only on thresholds.
- Learn what normal looks like for each hour and weekday.
- Link repeat incidents and fix the root cause.
- Keep change records in your monitoring timeline.
Predictive → Autonomous
- Pick one frequent failure you understand well.
- Automate the fix with a rollback and a log.
- Keep a person able to review and reverse every run.
- Measure what the automation does before adding more.
Autonomy is earned. Automate the failures you understand well, keep a human in the loop, and make every automated fix reversible.
Measuring the move: MTTx done right
- MTTD, mean time to detect
- How long it takes from the start of a problem until someone or something knows about it (our plain-words definition).
- MTTA, mean time to acknowledge
- The average time from when an alert is triggered until work begins on the issue.
- MTTR
- It can stand for four different measurements (repair, recovery or restore, respond, resolve), so agree on and write down which one you track.
- Agree what each metric means and write it down.
- Report medians and the slowest 10%, not just averages.
- Count who found it first: tools or people.
Illustrative. Twelve months of an invented operation moving from reactive to proactive: median time to detect falls from 95 to 30 minutes while the share of incidents found by tools first rises from 35% to 80%.
Solid line, left scale: median time to detect, in minutes (Illustrative). Dashed line, right scale: share of incidents found by tools first (Illustrative).
The posture check
Ten questions, about two minutes. You get a score out of 20, your place on the ladder, and steps to fix each gap. Your answers stay in your browser unless you choose to send a project request.
The check needs JavaScript.
Sources
Facts checked Oct 13, 2026.
Illustrative examples are invented to explain the ideas and are not measured data. Survey figures are as published by their sources; King Tech is independent and is not affiliated with any of them.