Skip to main content

Getting Started with Scheduling

The Scheduler is your platform's alarm clock for code. Tell it "run this at 9am every weekday" or "remind me in 30 minutes" and it handles the rest — retries on failure, logs every run, notifies you when something doesn't go through.

What you'll learn​

  • What you can schedule
  • The three ways to set up a schedule (one-shot, recurring, advanced)
  • How to watch your scheduled jobs run
  • What happens when something fails

What you can schedule​

Anything that can be triggered by a single action — typically one of:

ActionExample
Run a workflow"Generate the weekly metrics report every Monday"
Call a webhook"Ping the on-call rotation at 9am"
Send a notification"Remind everyone about the standup in 5 minutes"
Fire a custom actionAnything someone in your org built that exposes an action interface

You schedule the what once; the scheduler runs it as many times as your schedule says.

Three ways to set up a schedule​

Option 1: Run once at a specific time​

The simplest case. Pick a date + time. The job runs once and is done.

Example: "Send a launch announcement at 9am next Tuesday."

Option 2: Recurring — pick a frequency​

For the common patterns, you pick from a dropdown: every day, every Monday, every weekday, once a month, etc. No advanced syntax needed.

Example: "Run the backup every day at 2am."

Option 3: Advanced — cron expressions​

If you need something the dropdown doesn't cover ("every 15 minutes during business hours, only on weekdays"), the scheduler accepts cron syntax. Copy-paste from a cron expression generator if you're not familiar.

Example: 0 9 * * MON-FRI = 9am every weekday.

Create a job​

  1. Scheduler → New Job in the web app.
  2. Give it a name + description.
  3. Pick what to run (workflow, webhook, etc.).
  4. Pick when (one-shot / recurring / cron).
  5. (Optional) Set up notifications for success/failure.
  6. Create.

The job is active immediately — the scheduler will fire it the next time the schedule says to.

Watch jobs run​

Scheduler → My Jobs lists everything you've scheduled, with:

  • The next run time
  • The last few runs and whether they succeeded
  • A button to Run now if you want to trigger ad-hoc

Click any job to see its full run history (timestamps, durations, outputs, errors).

What happens when something fails​

A job that fails is automatically retried with exponential backoff. After enough failures, it moves to the dead-letter queue — that means we stopped trying, and a notification lands in your inbox.

To recover:

  1. Open the failed job.
  2. Look at the Last error field to see what went wrong.
  3. Fix the underlying issue (your webhook URL, the credentials, etc.).
  4. Retry from dead letter to give it another shot.

→ Troubleshooting

Pause / disable a job​

You can pause any job to stop it from running without deleting it. Useful for "we're freezing this for the next two weeks."

Disable is the same but permanent — the job stays in the list but won't fire until you re-enable it.

Next steps​