Your first monitor

The fast way

On the monitors page, use Quick add and give it a domain. It creates three monitors at once:

• an HTTP monitor on the site itself, so you know if it is up;
• an SSL monitor on its certificate, so it does not expire on a Sunday;
• a DNS monitor on its records, so you find out if they change.

Those three cover most of what goes wrong with a website, and it is the fastest way to get something useful running.

Creating one by hand

New monitor, then:

1. Type. What kind of check. See choosing a type if you are not sure — the choice decides which other fields you get.
2. Name. For you, not for the check. "Marketing site" beats "https://example.com/" when it appears in an alert at 3am.
3. URL or host. What to check.
4. Interval. How often. Your plan sets the floor; see settings.
5. Regions. New York is always included.

Save, and the first check runs within the interval. You can also press Check now.

Make sure the alert reaches you

This is the step people skip, and the one that matters.

With no notification channels configured, we email every member of the team. That is a deliberate fallback — monitoring that alerts nobody is worse than no monitoring, because you think you are covered. But it is a fallback, not a setup.

Go to Settings → Notifications, add a channel (Email, Slack, Microsoft Teams, Discord, Webhook), and send a test. The test is the point: a channel that was going to fail will fail now, while you are watching, rather than during the outage.

Then decide which monitors use it. A channel can cover everything, or only the monitors you attach it to.

Reading the result

The monitor's page shows its uptime, a response-time chart, and the recent checks with their outcome. See reading a monitor's page.

A new monitor shows Unknown until its first check finishes. That is not an error, it just has not looked yet.

What to do next

• Add the status page your customers can look at instead of emailing you.
• If you run your own servers, install the agent.
• Put a heartbeat on your backup job, which is the thing most likely to have been failing silently for months.