● Know within seconds, not days

Your cron jobs fail silently.
We make them speak up.

Backups, scrapers, billing scripts, and nightly reports break without a sound — and you find out when a customer does. Pulsewatch pings you the moment a scheduled job misses its check-in.

Pulsewatch monitors cron jobs and scheduled tasks. Add one ping URL to your job, and Pulsewatch alerts you by email or Slack if it stops checking in.

No credit card · 3 monitors free · Setup in one line

curl -fsS https://app.pulsewatch.io/ping/YOUR_TOKEN

Add that one line to the end of any job. Silence = we alert you.

The problem isn't that jobs fail. It's that they fail quietly.

Uptime monitors watch your website. Almost nothing watches the invisible work happening on a schedule.

💾 Backups that stopped weeks ago

You assume last night's dump ran. It's been failing since the disk filled up. You learn this during a restore.

🕷️ Scrapers & ETL that die overnight

A source changed, the cron exited non-zero, dashboards quietly went stale. No error reached a human.

🧾 Billing & report jobs that skip

The monthly invoice run didn't fire. Finance notices on the 5th. Now it's a customer trust problem.

How it works

A "dead man's switch" for anything on a schedule. If the check-in doesn't arrive on time, you hear about it.

1
Create a monitor
Tell us how often the job should run and how much grace to allow.
2
Add one line to your job
Ping your unique URL at the end of the script or as the last cron step.
3
We watch the clock
Ping arrives on time → green. Ping is late → we mark it down and alert you by email & Slack.
4
Recovery is automatic
Next successful ping flips it back to green and sends an all-clear — no manual reset.

Cron example — one extra step:

0 3 * * * /usr/local/bin/backup.sh \
  && curl -fsS .../ping/YOUR_TOKEN

Only pings on success (&&). If the backup fails or the box is down, no ping arrives — and we tell you.

Want duration & failure signals too?

curl .../ping/TOKEN/start  # job began
curl .../ping/TOKEN      # success
curl .../ping/TOKEN/fail  # job reported failure

Everything you need, nothing you don't

⚡ Instant failure detection

A background clock checks every monitor continuously. Late = down, no polling your servers required.

🔔 Email + Slack alerts

One alert when it breaks, one when it recovers. We only notify on real state changes — never spam.

🟢 Recovery detection

Know when it's fixed, not just when it broke. Full incident history per monitor.

🔒 Unique, revocable tokens

Every monitor gets its own 144-bit URL. Isolated per account. Rotate anytime.

🧩 Works with anything

Cron, systemd timers, GitHub Actions, Kubernetes CronJobs, Lambda, Windows Task Scheduler, bots.

🚀 60-second setup

Create a monitor, copy one curl line, done. No agent to install, no SDK required.

Stop finding out from your customers.

Set up your first monitor in the next two minutes — free.

Create your first monitor →

Pricing

Public monthly pricing. Free forever for personal jobs — upgrade as you add more monitors.

Free

$0/month

3 monitors

Good for testing and personal jobs

Starter

$5/month

25 monitors

For solo developers and small projects

Pro

$12/month

100 monitors

For indie SaaS founders and small teams

Team

$29/month

500 monitors

For agencies and production teams

Taxes may apply and will be calculated at checkout.

FAQ

How is this different from uptime monitoring?

Uptime tools ping your server from outside. Pulsewatch works the opposite way: your job pings us. That's the only way to catch a job that silently didn't run at all — there's nothing to poll.

What if my server has no internet / is behind a firewall?

A single outbound HTTPS call is almost always allowed. It's one request; if even that can't get out, that itself usually signals a bigger problem worth an alert.

Will I get spammed?

No. We alert once on the transition to "down" and once on recovery. A job that stays down does not re-notify.

Do you support failure & duration signals?

Yes — ping /start then / to measure run time, or /fail to report an explicit failure and trigger an immediate alert.

Is my data isolated?

Every monitor belongs to one account and is only reachable via its unguessable token. Passwords are salted & hashed (scrypt).

Can I self-host?

Yes. Pulsewatch is a single Node process with an embedded database — no external services required to run the core product.