Find out that a cron job never ran before your customers do
Regular monitoring tells you the server is alive. It will not tell you the nightly backup has not started for three weeks. Tickwatch watches your job schedules — and shows the result to your customers on a public page.
- 5 monitors free forever
- no card required
- 5 minutes to set up
The server is green. There has been no backup for three weeks.
Uptime monitoring checks that a port answers. It has no idea you even have a nightly backup — let alone that it quietly stopped running.
And it says nothing about your jobs.
not when a customer asks you to restore their data.
Five minutes from signup to your first alert
Create a monitor
Describe the schedule — a cron expression, an interval, or on demand. Time zone and grace period are configured separately.
Add one line
The job reports in with a plain HTTP request. Works from anywhere — nothing to install or configure on the server.
Hear only what matters
Silence from a job is the signal. We work out when the ping should have arrived and write to you only if it did not, or arrived with an error.
If it can make an HTTP request, it is already compatible
No SDKs, no agents. One curl at the end of the command and the job is on the map.
The status page competitors charge extra for
A public page on your own domain: components, incidents, maintenance windows and subscriptions for your customers. Powered by the same monitors — no manual upkeep.
This incident was created automatically: the “Warehouse sync” monitor did not report within its expected window. Updates will appear here and go out to subscribers.
The page is served from a pre-computed snapshot and lives apart from the dashboard — it has to keep working precisely when something has gone wrong on our side.
Notifications you will not want to mute
Monitoring dies from too many alerts, not too few: after two weeks of noise the team simply stops reading the channel. That is why fighting noise here is part of the product, not a setting.
- Schedule
- 0 3 * * * (Europe/Moscow)
- Exit code
- 1
- Duration
- 4 min 12 s
The tail of stderr right inside the notification is usually enough to understand the cause without opening a laptop.
Priced in roubles, paid by card, with proper invoices
No exchange-rate maths and no surprises with cross-border payments.
- 5 monitors
- 7 days of history
- 1 status page
- Email and Telegram
- Status page on our subdomain
- One member
- 25 monitors
- 30 days of history
- 1 status page
- Custom domain for the status page
- Slack, Discord, webhooks
- Up to 3 members
- 100 monitors
- 90 days of history
- 3 status pages
- SMS, 100 per month
- SLA reports
- Up to 10 members
- Priority support
- 500 monitors
- 365 days of history
- 10 status pages
- SSO and audit log
- On-call schedules
- Unlimited members
- Contracts and invoices
Hitting a limit blocks creating a new monitor, but never switches off the ones already running. Breaking your monitoring over a late payment is the last thing we would do.
What people usually ask
Uptime monitoring answers “is the service responding?”. We answer “did the job actually run?”. Those are different questions: a server can be perfectly healthy while the nightly backup has not run for weeks because someone deleted a line from crontab.
No. The job makes one HTTP request when it finishes. We need no agent, no daemon and no access to your infrastructure — we never connect to your servers at all.
Nothing. The next run is computed in the monitor's own time zone by parsing the cron expression, not by adding 24 hours. That is exactly why nightly jobs do not produce false alarms twice a year — a day with a clock change lasts 23 or 25 hours, and we account for it.
That is what the grace period and the separate duration limit are for. While the job fits inside the grace period nobody is disturbed. If it started and then hung for longer than allowed, that is a distinct kind of alert, because the cause is usually different.
Ping ingest and status pages run separately from the dashboard, and our scheduler pings an independent external service. If the scheduler stops, we hear about it — not you. Monitoring that quietly dies together with its customer is worthless.
Card payments in roubles. The receipt is issued automatically. For companies we prepare invoices and closing documents — billing details live in the workspace settings.
Monitors and notifications that already run keep running — we do not switch alerts off over a late payment. Only creating new ones is blocked, and limits drop to the free plan. Data is kept for another 30 days and can be exported.
Your first monitor — in five minutes
Five monitors free forever, no card. If it does not work out in ten minutes, write to us and we will sort it out together.