Plays nice with the tools you already run
The actual product
Not a mockup. This is the app.
Every screen below is the real product, captured live. Click through the incident lifecycle — from the first page to the AI root-cause to the customer status page.
The 2am problem
Right now, an outage means chaos.
Most teams run incidents across five disconnected tools — the monitor that saw it, the pager that woke someone, the Slack thread where it got debugged, the status page nobody updated, and the postmortem nobody wrote. Five tools, no owner. The same three gaps open every time.
The wrong person wakes up
The on-call rota lives in a spreadsheet nobody updated. The alert goes to someone on holiday. Twenty minutes pass before a human even sees it.
Nobody's running the room
Three people debug the same thing in three DM threads. Leadership asks "what's the customer impact?" and nobody can answer.
The lesson is lost
The postmortem never gets written. Six weeks later, the exact same Redis outage takes you down again — and again.
The complete platform
Everything from the first
alert to the final postmortem.
The work of PagerDuty, Statuspage, Pingdom, and your postmortem template — designed as one system, not four subscriptions. Everything below is shipped and running in production today.
The triage pipeline
The thing the product is named for. We dedupe noisy alerts, correlate the ones that share a cause, and hold low-confidence signals in a silent Pending gate until they self-resolve or recur. You're woken for signal — not noise. No competitor ships this.
The wedgeOn-call & escalation
Visual schedules with rotations and one-tap overrides, plus multi-layer escalation. We page by phone, SMS, Slack, email, and push — and keep climbing the chain until someone answers.
AI investigation
Click investigate and the AI weighs the likely causes in parallel, commits to a verdict with a confidence score, and drafts the code fix as a reviewable PR — every claim backed by evidence, nothing shipped on its own.
Ask triage
Ask production anything in plain English — no incident required. Read-only across your connected sources, scoped to your stack by the System Map, every answer backed by evidence you can open.
NewUptime & status pages
Multi-region HTTP, TCP, ping, and cron-heartbeat checks, plus SSL and domain-expiry warnings. Ships with branded public status pages your customers can subscribe to.
Incident response in Slack
Declare an incident from Slack and we spin up a war room, assign a commander, and keep a live timeline of every decision. Run the whole thing without leaving Slack.
Postmortems, written for you
When it's over, AI drafts the blameless postmortem from the timeline. During the incident it also suggests severity and surfaces similar past incidents — each grounded and one click to accept.
Change attribution
We pin what changed right before the incident — "fired 90s after deploy 8f3c2a1 by Marco" — from Vercel, GitHub Actions, and LaunchDarkly. No more "what did we ship?"
Native mobile paging
A real iOS & Android app that rings through Silent and Do Not Disturb, throws a full-screen page over the lock screen, and lets you acknowledge with one tap.
NewAI investigation
AI that finds the cause —
then drafts the fix.
When an incident opens, the AI does what a senior engineer would: weighs the likely causes in parallel, grounds each one in your live telemetry, and commits to a verdict — then proposes the exact code change to fix it.
- Weighs hypotheses in parallel — ranks the likely causes by confidence and commits to a verdict, each one backed by the evidence behind it.
- Drafts the actual fix — proposes the code change as a pull request you review and merge. Every action is a human click — it never ships on its own.
- Ask a follow-up — keep questioning the investigation in plain English; every answer cites evidence you can open.
- Reads your live telemetry — Prometheus, Loki, Datadog, CloudWatch, Tempo traces, Kubernetes, and GitHub, with secrets masked before anything is sent.
Included — not an add-on
Your AI, not an AIOps tax.
Bring your own AI: the investigator runs on your own Claude, Gemini, or local model key — so it's part of every seat, never a separate AIOps SKU or add-on bill. It plugs into the telemetry you already run — Prometheus, Datadog, CloudWatch, Loki, Tempo, Kubernetes, and GitHub — and because it's your key and your model, raw data never leaves your network.
Ask triage · new
Ask production anything.
No incident required.
Talk to your whole stack in plain English — read-only across every connected source, scoped to your services by the System Map. Every answer is backed by evidence you can open. No incident, no dashboards, no query language.
- No incident required — ask any time: “what changed on payments-api?” or “is checkout latency elevated right now?”
- Scoped by your System Map — answers are bounded to your real services and their dependencies, not guesses about the whole internet.
- Evidence you can open — every claim cites the file, metric, or deploy behind it, and unverified references are flagged.
- Read-only and safe — it inspects, it never changes anything, with the same masked-secrets guarantee as the investigator.
On-call, done right
A rota everyone can actually see.
Build a rotation in a couple of clicks. Swaps and overrides take seconds. And the right person is always one tap away from acknowledging.
- Follow-the-sun rotations — daily, weekly, or custom handoffs across timezones, DST-correct.
- One-tap overrides — cover a teammate's shift without a spreadsheet edit war.
- Escalation that never stops — if no one answers, it keeps climbing the chain.
The app
Carry the pager
in your pocket.
A real native app for iOS and Android — not a notification you'll sleep through. The paging core is native, so a SEV1 rings even when your phone is silenced.
- Rings through Silent & Do Not Disturb via Critical Alerts — the way a pager should.
- Full-screen page over the lock screen, with a built-in reliability checklist so you know it'll get through.
- Acknowledge from the lock screen — one tap, no unlock, full incident parity with the web.
The app is yours the moment you sign in — grab the Android build in-app. iOS via TestFlight shortly: it's built and waiting on Apple's Critical Alerts entitlement.
Pricing
€25 per user. Everything in.
One price per person for the whole platform — monitoring, on-call, AI investigation, status pages, and postmortems. Every feature, every integration, included. Cheaper per seat than tools that do a fraction of it.
| Tool | Per user / month | What you actually get |
|---|---|---|
| PagerDuty Business | ~€38 | Paging only |
| incident.io + On-call | ~€42 | Incident response + on-call |
| Rootly | ~€28 | Incident response, sold in modules |
| triage.events | €25 | The whole platform |
- Monitoring, on-call & escalation
- AI investigation + postmortems
- Status pages, SLOs & workflows
- SSO, audit log & every integration
€25 per user, per month, backed by a 14-day money-back guarantee — no questions asked. Enterprise is custom, built around your security and procurement.
On the clock
Leaving OpsGenie?
Atlassian is sunsetting OpsGenie in 2027. This time, get the whole platform — not just another pager. Move to Triage in 30 minutes — our importer brings your schedules, escalation policies, and services across, with a parallel-run mode so you can verify before you cut over.
"I've spent years on-call at scale, with a whole platform team running the tooling that made it bearable. Most teams never get that. Triage is that entire stack — monitoring, paging, investigation, comms, postmortems — built to run itself, so a ten-engineer startup responds like it has a hundred."
The next incident is already on its way. Be ready.
Start free in under a minute — we wire your first integration and page you, so you see it work before you pay a cent.
Free for 14 days. No credit card, no sales call.