Vigil is the on-call command center for SaaS teams. Smart routing wakes the right human, AI drafts the timeline, and your status page stays honest — automatically.
Burn-rate monitor paged payments-primary via app + call.
“Looking — rolling back canary first.” War room opened.
Correlated deploy pay-api@8f3c1a + 214 related errors.
Rollback to 8f3b77 would restore p99 in ~6 min. Drafted status update + customer notice. Approve →
On-call teams at fast-moving companies run on Vigil
Routing, schedules, war rooms, status pages and learning loops — connected so nobody pastes links at 3am.
Urgency-aware rules combine service ownership, past fixes and who's actually awake. No more paging the whole team.
Follow-the-sun, overrides in two taps, and load scores that flag burnout before it pages someone.
Slack channel, bridge, timeline and roles spawn together. Guests get scoped access automatically.
Incident state publishes itself. Pre-written templates keep wording calm, clear and consistent.
Timelines, graphs and action items are drafted from the response — you just add the lessons.
Three modes, one rail. Switch tabs to see what Vigil does at each stage.
200 related alerts collapse into a single incident with severity, blast radius and likely cause attached.
# 214 alerts → 1 incident group_by: service + deploy_sha correlate: errors ↑ 18x after pay-api@8f3c1a severity: SEV-2 (3.1% checkouts) page: payments-primary (Priya · 41s median)
Commander, comms and ops roles are assigned on open. Runbooks and rollbacks sit next to the timeline — not in another wiki.
/vigil role commander @priya /vigil page payments-secondary /vigil runbook checkout-latency-v4 status → “Identified — rolling back canary” (auto-posted)
Timelines write themselves from pages, chats and deploys. Action items open as tracked tasks with owners.
## What happened (drafted) detected 02:14 · mitigated 02:31 · resolved 02:58 ## Actions → [ ] canary auto-halt on burn-rate @sre [ ] p99 alert threshold 800→600ms @payments
Escalations, pages and status updates as readable rules. Humans approve the moments that matter.
on: burn_rate > 4x for 2m do: - page: payments-primary (app + call) - open: war-room + bridge - draft: status-page "Investigating" escalate: +secondary after 8m unacked resolve: auto-close when burn < 1x for 15m
“We cut pages by two-thirds in a month. The first SEV after switching felt… boring. That's the highest praise I can give.”
“Customers stopped asking 'is it down?' Our status page updates itself before support even notices. Trust went up, tickets went down.”
Unlimited alerts and status pages on every plan. 14-day trial, migrate from PagerDuty in an afternoon.
Routing is only the start. Vigil groups related alerts into one incident, drafts timelines and status updates with AI, and closes the loop with postmortems — so a page becomes a 20-minute fix, not a 3-hour thread.
Most teams import schedules, services and escalation policies in under an hour with our PagerDuty / Opsgenie importer, then run both in shadow mode for a week before switching.
Yes. Load scores show nights paged, interruptions per rotation and override debt, so managers can rebalance before someone burns out.
Pages fail over across regions and to SMS/voice carriers directly. Our status page is hosted on fully separate infrastructure — 99.99% delivery over the last 12 months.
No. Team and Enterprise workspaces are never used for model training. Ephemeral AI sessions, scoped tokens, SOC 2 Type II and EU residency available.
Import your schedules today. Page yourself tonight (safely) and watch Vigil route, draft and resolve.
No credit card · Cancel anytime · Free PagerDuty import