Know the moment your website goes down
Monitor websites, APIs, SSL certificates, domains and cron jobs. Get alerts by email, Slack, Discord, Telegram or webhook, and keep customers informed with a public status page.
- Free plan, no credit card
- Checks every 30 sec
- Set up in 2 minutes
Uptime, 30 days
99.98%+0.02%
Avg. response
142 ms-8%
Open incidents
0
- Websitehttps://example.com99.98%
- Checkout APIapi.example.com/v1100%
- Nightly backup0 2 * * *99.91%
- SSL certificateexample.com:44354 d
- monitor types
- 7
- alert channels
- 6
- fastest check interval
- 30 sec
- of uptime history
- 90 days
Alerts delivered to
- Slack
- Discord
- Telegram
- Microsoft Teams
- Webhook
Monitoring
Seven ways to catch problems before your customers do
Every monitor feeds your alerts, incidents and status pages automatically.
Website & API
HTTP(S) checks with expected status codes, custom headers and response-time limits.
GET /health → 200 · 142 ms
Keyword
Make sure a page still says what it should, or get alerted when an error message appears.
contains "In stock"
TCP port
Watch databases, mail servers, SSH and anything else that listens on a port.
db.example.com:5432
SSL certificate
Get warned weeks before a certificate expires, or the moment it becomes invalid.
example.com:443 · 54 d
Domain expiry
Never lose a domain name to a forgotten renewal again.
example.com · 212 d
DNS
Check that your records resolve and still point where they should.
A → 203.0.113.10
Cron jobs & heartbeats
Backups, queues and scheduled jobs check in with you. Silence means trouble.
0 2 * * * /api/ping/…
Every type, on every plan
The free plan included. Plans differ in limits, never in monitor types.
Start monitoring free
Alerts
Alerts where your team already works
Send each monitor to as many channels as you like. Failures are confirmed before anyone is paged, and you hear the moment things recover.
Email
To you and your whole team
Slack
Rich messages in any channel
Discord
Straight into your server
Telegram
Private chats and groups
Microsoft Teams
Cards in a team channel
Webhook
Signed JSON for your own tools
Check failed: connection timed out
Confirming, 1 of 3Confirmed down, alerts sent
SlackEmailMicrosoft TeamsIncident opened on your status page
Subscribers notified by emailStill down reminder
SlackRecovered after 35 min
Incident resolved automatically
- No false alarmsA monitor is only marked down after several failed checks in a row.
- Still-down remindersLong outages are never forgotten: get reminded until they are fixed.
- Quiet during maintenanceScheduled maintenance windows mute alerts and keep visitors informed.
Status pages
A status page your customers will actually check
Publish live uptime, incidents and maintenance on a branded page, and answer "is it just me?" before it becomes a support ticket.
- Your brandLogo, accent colour and a light or dark theme.
- Email subscribersVisitors subscribe and get every update.
- Incident historyUp to 90 days of uptime bars and a full archive.
- RSS & Atom feedFollow updates in any feed reader.
- Status badgesLive SVG badges for READMEs and docs.
- Embeddable widgetShow live status inside your own app.
- Website100% uptimeOperational
- Pricing page99.95% uptimeOperational
- SSL certificate100% uptimeOperational
- Public API99.97% uptimeOperational
- + 3 more
Cron job monitoring
Find out when a scheduled job silently stops
Add one line to your crontab, backup script or CI pipeline. If the ping doesn't arrive on time, you're alerted.
- …/api/ping/YOUR_TOKEN
- The job finished. Resets the timer.
- …/api/ping/YOUR_TOKEN/start
- The job started. Measures how long it runs.
- …/api/ping/YOUR_TOKEN/fail
- The job failed. Alerts right away.
# Back up the database every night, then check in
0 2 * * * /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 3 https://statuszen.vercel.app/api/ping/YOUR_TOKENHow it works
Up and running in three steps
- 1
Add what to watch
Paste a URL, host or port, or create a heartbeat for a cron job. Checks start right away.
- 2
Choose who hears about it
Connect email, Slack, Discord, Telegram, Microsoft Teams or a webhook in a few clicks.
- 3
Share your status
Publish a status page so customers see what's happening without having to ask.
Pricing
Start free, upgrade when you grow
Every plan includes all monitor types and alert channels.
- Free$0foreverMonitor your most important website.Monitors5Check interval5 minStatus pages1
- RecommendedPro$15per monthFor businesses that can't afford downtime.Monitors50Check interval1 minStatus pages5
- Agency$39per monthFor agencies and teams monitoring many clients.Monitors500Check interval30 secStatus pages50
FAQ
Questions, answered
Everything you need to know before you add your first monitor.
Where do checks run from?
From our monitoring servers, around the clock. A monitor is only marked down after repeated checks confirm the failure, so a single network blip won't wake you up.
How often are my monitors checked?
As often as every 30 sec, depending on your plan. Heartbeat monitors follow the schedule and grace period you choose.
Do I need to set up a cron job?
No. Checks can run in three ways: a small background worker, any cron service that calls a secret URL, or automatically while the site receives visitors. Use whichever your hosting supports.
Can I monitor internal servers?
Private and internal addresses are blocked by default, so the service can't be used to probe private networks. On a self-hosted install the administrator can allow them, for example to watch servers on a company network.
Will I get an alert for every small blip?
No. You choose how many failed checks in a row confirm an outage. You get one alert when a monitor goes down, optional reminders while it stays down, and a recovery notice with the total downtime.
Can my customers subscribe to updates?
Yes. Visitors can subscribe by email or follow the RSS and Atom feed, and you can embed a live status widget or badge in your site, app or docs.
From the blog
All articlesOct 16, 2026
Incident communication: what to post on your status page
Clear, timely updates turn an outage into a moment of trust. A simple template for every stage of an incident.
Read articleOct 15, 2026
How to monitor cron jobs and backups with heartbeats
A failed backup is silent until you need it. Heartbeat monitoring tells you the moment a scheduled job stops running.
Read articleOct 14, 2026
What is uptime monitoring and why every website needs it
Your customers notice downtime before you do — unless something is watching for you. Here is how uptime monitoring works.
Read article
Be the first to know when something breaks
Add your first monitor in two minutes, then get back to building. We'll keep watch.
Free plan, no credit card needed.