Incident escalation policies that page until someone answers.
The alert that wakes nobody is the one that costs you the night. If the first responder doesn't acknowledge, Uptimia pages the next person, then the next — automatically, until someone picks it up.
Dry run
example incident · recomputed as you editThe page nobody answered
02:14 on a Tuesday: checkout stops working and the incident is confirmed. The on-call engineer is asleep, the phone is face down, and five minutes pass with no reply.
One ladder, every monitor type
One escalation policy attaches to all 12 monitor types — uptime, SSL, transaction, DNS, heartbeat, API and the rest. Every confirmed incident on them climbs the same tiers.
Build the ladder once, reuse it everywhere
Name a policy, add timed steps, and pick who each one pages — from your teams, people and individual contacts. A live dry run shows exactly who gets reached before you save.
- Up to 10 tiers per policy — page these people now, these in 5 minutes, everyone in 15
- Any recipient — teams, individual members, operators or single contacts, up to 30 per step
- A live dry run resolves every step to real people, and warns about a step that would reach no one
Climb, acknowledge, recover
One tap acknowledges — no login required
Pages sent to email, SMS, Slack, Microsoft Teams, Discord, Mattermost, Telegram and Twilio carry a signed acknowledgement link. Tap it from the alert — no dashboard, no password — and the first acknowledgement stops the climb for everyone on the incident.
- One tap from the alert — a signed link, no login and no app to open
- The first acknowledgement stops the climb — nobody further up the tiers is paged
- Time to acknowledge recorded on every escalated incident, so response time is measured, not guessed
How long until someone answered
"We respond fast" is a hope until a number sits next to it. Every escalated incident is stamped with the minutes between confirmation and the first acknowledgement, and with the name of whoever answered.
- Minutes to acknowledge — counted from the moment the incident was confirmed to the first acknowledgement
- Who answered, and from where — the control panel, or the signed link in the alert itself
- Unanswered incidents say so — the history row reads "recovered before step 2", naming where the ladder stopped
Status, next step and MTTA
The incident page opens with a stat strip built for the middle of an outage: whether the ladder is still escalating, what time the next step fires, who acknowledged and how long that took.
- Escalating, Acknowledged or Resolved — the state of the ladder as a single badge
- Next escalation with a countdown — "Step 3 · 5 min", or "Paused" once someone acknowledges
- MTTA in minutes — confirmation to first acknowledgement, on the same strip
All-clear only to who was paged
When the monitor recovers, the ladder cancels every pending page at once — and the recovery notice reaches only the people it actually alerted, not the wider team it might have paged three tiers later.
- Recovery cancels every pending step and repeat the instant the monitor is back
- Only the recipients it paged get the all-clear — nobody else is notified
- Quiet hours honoured at every tier, per person — a ladder never overrides them
The next 3 a.m. incident should wake the right person.Not everyone.
On-call ladders without on-call pricing — escalation, acknowledgement and team seats come in the flat plan price, never billed per user.
How escalation policies work
Build a ladder once, attach it to your monitors, and let an unanswered incident climb it on its own.
Build the ladder
Add timed steps — page these people now, these in 5 minutes, the whole team in 15 — choosing recipients from your teams, members and contacts.
Attach it to monitors
Attach the policy to any monitor — uptime, transaction, SSL, heartbeat. One ladder can cover the whole fleet; a monitor with no policy simply alerts everyone at once.
It climbs until someone answers
When an incident is confirmed, step 1 fires immediately. If no one acknowledges, the ladder pages the next tier on schedule — until someone acknowledges, or the monitor recovers.
In the plan price, never per seat
Up to 10 tiers per policy
Chain up to 10 timed steps, each paging teams, members or individual contacts. A bounded repeat then re-fires the last step — or all of them — on a cadence from 10 minutes to 2 hours, for as many rounds as you allow, and stops there.
One-tap acknowledgement
A signed link in the alert acknowledges the incident — no dashboard, no password.
Time to acknowledge
Every escalated incident records the minutes to the first acknowledgement, and who gave it.
Pause silences the ladder
Pausing a monitor cancels its active escalation and clears every pending page.
Maintenance windows
A scheduled window suppresses alerts, so planned work never starts a ladder.
One policy across every monitor type
Attach the same ladder to uptime, SSL, transaction, DNS and heartbeat monitors — reused across the fleet, not rebuilt per monitor.
Every tier, all 12 channels
A ladder pages through the same contact list and the same 12 channels every Uptimia monitor already uses — no second directory to maintain.
Microsoft Teams12 channels, one contact list — every tier of a ladder pages through the same routing.
Browse the full integrations directory →What is an incident escalation policy?
An incident escalation policy is a reusable, timed ladder of alert steps: it pages the first responder the moment an incident is confirmed, and if no one acknowledges, it automatically pages the next people on a schedule you set — until someone answers or the monitor recovers. It turns a single fire-and-forget alert into an accountable chain, so an incident is never left sitting in one unread inbox.
How does an escalation policy work?
Step 1 fires the instant the incident is confirmed. An acknowledgement stops the ladder immediately, and no further tier is paged.
Climb, then widen
no acknowledgement in 5 min → the ladder pages the next tier, then repeats until someone answers
All at once versus a ladder
Alerting everyone at once works — until the one person who can fix it is asleep, off-shift, or has muted the channel. A ladder pages the right person first, then widens only if nobody answers.
See alerting & notifications →| What happens | Alert all at once | Escalation ladder |
|---|---|---|
| Incident is confirmed | ✓ Paged | ✓ Paged |
| First responder doesn't answer | Silence | ✓ Next tier paged |
| On-call is asleep at 3 a.m. | Missed | ✓ Climbs to backup |
| Incident stays open | No repeat | ✓ Repeats, bounded |
| Monitor recovers | All-clear to everyone | ✓ Only to who was paged |
Incident escalation FAQ
01What is an incident escalation policy?+
02How is this different from alerting everyone at once?+
03What is one-tap, no-login acknowledgement?+
04What happens if nobody acknowledges?+
05Who gets the recovery notice?+
06Can one policy cover many monitors?+
07Does escalation respect quiet hours?+
08What if I delete a policy, or a step has no one in it?+
09Does pausing a monitor stop its escalation?+
10Can the ladder repeat forever?+
11Which channels can a step page?+
12Which plans include escalation policies?+
Reach someone who can fix it
Build a reusable escalation ladder in minutes, attach it to every monitor you own, and stop hoping the first alert was the one that got through.