Tallwatch
Back to blog
2 min read
IncidentsOn-call

What acknowledge means on an incident (and why it matters)

Ack is not resolve. It means a human owns the problem, and it should stop the escalation clock from climbing further.

NK

Nabin Khair

Founder

What acknowledge means on an incident (and why it matters)

Incident tools put an Acknowledge button next to Resolve. Teams mash them interchangeably. Then escalation policies misbehave and postmortems cannot tell when a human actually engaged.

Acknowledge

Someone responsible is awake to the incident and taking point. It should:

  • Stop escalation from climbing to the next level
  • Show on the timeline for MTTA (mean time to acknowledge)
  • Not close customer impact

Ack is a promise to work the problem, not a claim it is fixed.

Resolve

Customer impact is gone (or mitigated enough that you are no longer in incident mode). Monitors may auto-resolve when checks recover; humans still own the public message.

Why the distinction matters

If ack does not stop escalation, backups get paged while primary is already debugging: noise and resentment.
If people "resolve" to silence their phone while the site is still broken, you lose the incident record and the status story.
If nobody acks, you cannot tell detection problems from response problems in MTTD vs MTTR.

Hygiene

  • Ack when you open the laptop for that incident, not when you glance at Slack.
  • If you ack and hand off, say who is taking it.
  • Do not ack "for the team" from a shared account without a name.

Acknowledgment belongs in the incident lifecycle so escalation respects a human claiming ownership. Use the button that way: ownership signal first, green banner later.

Related

Keep reading

False alerts and status pages.

How to explain downtime to non-technical founders and investors

How to explain downtime to non-technical founders and investors

They do not need Kubernetes. They need impact, duration, cause at the right altitude, and what changes so it is less likely next time.

NKNabin Khair
When a CDN outage is your outage (even if origin is fine)

When a CDN outage is your outage (even if origin is fine)

Customers do not care that your origin returned 200 on a private path. If the CDN is how they reach you, its bad day is your incident.

NKNabin Khair
How long should an incident stay open after the site recovers?

How long should an incident stay open after the site recovers?

Auto-resolve is not the same as done. Close when customer impact is gone and the next on-call would not be confused, not the second a single check turns green.

NKNabin Khair