Three subscriptions
A status page product, a monitoring product and an on-call product, each with its own logins, its own bill and its own idea of who is on call.
Team gives a product team of up to 10 people one place for the checks, the status page customers read, and who gets woken at night: 100 monitors, escalation and on-call schedules, deploy markers and AI drafts of incident updates.
On Team, $39 a month.
Everyone on call, no fee per person.
Sign up, log in, the paths that matter.
An incident this soon after a deploy names it.
A status page product, a monitoring product and an on-call product, each with its own logins, its own bill and its own idea of who is on call.
Someone has to notice, open another tool and write an update while they are also fixing the problem.
The first question in every incident, usually answered by scrolling through CI history.
Write an incident report with the services it affects and how badly, then post updates as you go: investigating, identified, monitoring, resolved. Each update is emailed to the page’s subscribers once. A failing check marks its service on the page without anyone typing. Resolved reports stay in the history with an optional postmortem.
Checkout has worked normally for 20 minutes. We are sorry for the trouble.
We rolled back the release from 14:02. Payments are going through again and we are watching.
The errors started with a release at 14:02. We are rolling it back.
Some customers see an error when they pay. We are looking into it.
An escalation policy is a list of steps: who to notify and after how many minutes. A step can point at an on-call schedule, a daily, weekly or hourly rotation with overrides, so the alert reaches whoever holds the shift. Acknowledging from the push, the email or the chat message stops the next step.
CI reports each deploy with one request. It is marked on the response-time charts, and an incident that starts within 30 minutes of a deploy names it. When you write the status update, Claude drafts it from the incident’s facts: the affected services, when it started and how long it has lasted. You edit it and publish; nothing goes out on its own.
Send any method with your headers and body, expect the status codes you choose, and read a value from the JSON response. An API that answers slower than its limit, 2 seconds unless you change it, counts as down. TCP, UDP, DNS, ping and mail server checks cover the rest, and heartbeat monitors catch a cron job that stopped running.
JSON "checks.db" is "degraded", expected "ok"
Team covers 100 monitors, 10 team members and 10 browser flows, with 90 days of latency history and uptime kept forever. Pro has the same status page, escalation, deploy markers and AI drafts for 25 monitors and 3 team members, if you are smaller.
Compare every planFor a freelancer or a small team with a few sites.
Billed monthly, cancel any time
Get ProFor product teams that run a few services.
Billed monthly, cancel any time
Get TeamMonthly prices in USD, excluding VAT. Yearly billing costs less; see the pricing page.
For alerts by push, email, SMS, phone calls, Slack, Teams, Discord, Telegram and webhooks with escalation, schedules and acknowledgement, no. SMS and calls reach US and Canadian numbers for now and run on credits: Team includes 50 a month, with prepaid packs on top. If your on-call people are elsewhere, keep phone alerts where they are for now.
Start free, set up the checks, then pick Team under Settings → Plan.
Start for free