OpsDuty routes every alert to the right on-call engineer, escalates automatically when nobody responds, and keeps your team accountable after the incident.
No credit card required • Free tier for small teams
Getting the right person on the problem quickly is what separates a short blip from an outage. OpsDuty makes that path deliberate instead of accidental.
Routing rules match each incoming event against your own conditions and send it to the service that owns it, so alerts reach the team that can actually fix them.
If the first responder does not acknowledge in time, the escalation policy moves on to the next person automatically, across email, SMS, push, and phone calls.
Every incident keeps a full timeline of events, notifications, and notes, so the follow-up conversation is grounded in what actually happened.
From the first signal through to the post-incident review, in one place.
Build rotations from layers, with recurring shifts, rolling handovers, and per-person overrides when somebody swaps a weekend. Everyone sees the schedule in their own timezone.
Escalation policies decide who is notified, how, and how long they get to respond before the alert moves on. Notifications go out over email, SMS, mobile push, and phone calls.
Cron jobs and batch pipelines check in with OpsDuty on a schedule. When a check-in is missed, you get an incident instead of finding out days later that the job stopped running.
Give customers or partners a phone number that always reaches whoever is on call right now, with the call forwarded straight to them and logged against the incident.
Receive events by webhook or email from any monitoring system, and parse the payload yourself with templates. Incidents are mirrored into Slack, where you can also raise one or check who is on call.
Track response and resolution times over time to see whether on-call is actually getting better, and keep customers informed with a public status page.
OpsDuty does not need you to replace your monitoring stack. Anything that can send a webhook or an email can open an incident, and you control how the payload is read.
Start freeIncidents mirrored into channels, with slash commands to raise one or check who is on call.
Ingest events from any system, with a payload template you control.
Give any alerting tool an address and turn its mail into structured incidents.
A documented REST API with OAuth2 tokens, plus mobile apps with push notifications.
OpsDuty speaks the Model Context Protocol, so the assistant your team already uses can read your operational state and act on it. Connect once with OAuth2; it only ever sees what that person can already see in OpsDuty.
"Who is on call for payments right now, and who takes over tonight?" It follows the service's escalation policy down to the rotation behind it, including whoever is covering a shift.
Summarise what happened from the timeline, then acknowledge, note, escalate or resolve. Anything that pages a person is marked as such, so your client can ask you first.
Claude Code, Claude Desktop, Cursor and Codex have copy-paste setup inside OpsDuty, and anything else that speaks MCP over HTTP works the same way.
Start on the free tier with no credit card. Upgrade when your team outgrows it.
€ 0 To try things out
€ 9 Per seat per month
€ 17 Per seat per month
Custom Talk to us
No. The free tier lets you create an organization, set up a schedule, and receive real incidents without entering payment details. You only add a card when you move to a paid plan.
Over email, SMS, mobile push notifications, and automated phone calls. Each person sets their own notification policy, so you can escalate from a quiet channel to a loud one as an incident ages.
Yes. Anything that can send a webhook or an email can create incidents in OpsDuty. Incoming payloads are parsed with templates you control, so you decide which fields become the summary, severity, and deduplication key.
It catches the jobs that fail by not running at all. Your cron job or batch pipeline pings OpsDuty each time it completes, and if an expected check-in does not arrive, OpsDuty opens an incident.
Yes. SAML2 single sign-on and SCIM2 user provisioning are available on the Standard and Enterprise plans, so accounts can be managed from your existing identity provider.
Yes. Stakeholder accounts give non-responders visibility without a full seat, and public status pages let you communicate with customers during an outage.
Create an organization, add a schedule, and point your first alert at OpsDuty. The free tier is enough to see the whole flow end to end.
Start free