How do I check an automation's run history with the Events panel?

Open the Events panel

1. Go to Automations in the left navigation. The Events column shows how many times each automation has run. A red alert icon next to the count means the automation hit errors in the past 72 hours — hover over the icon and click View errors to open those events directly.

2. Open the automation you want to inspect.

3. Click Events at the bottom of the editor. The panel slides up over the canvas. Drag its top edge to resize it; click Events again (or the X on the right) to close it.

If the panel shows No events, the trigger has never fired. Confirm the automation is published and that the thing it listens for — a task completing, a placement being created — has actually happened. To fire the trigger yourself on safe data, see 📄 How do I test an automation before publishing?

Read an event

The panel has three sections, left to right.

4. Select an event in the Events list. Each row shows a status icon, the time it fired, and the candidate it ran for. Hover over a row for the exact timestamp. Use the status dropdown to filter the list, and Load more to page through older events.

5. Select a block in the Blocks list to focus on one step of the run. The matching block highlights on the canvas above, so you can see exactly where in the flow you are. All blocks shows the whole run end to end.

6. Work through the tabs on the right:

   • Logs — a running account of what each block did, in order, with successes in green and failures in red.

   • Data — the data the run evaluated. With All blocks selected, you see the data the event started with. With one block selected, you see a side-by-side comparison of the data before and after that block ran.

   • Trace — a timeline of chained runs, shown when this automation was triggered by (or itself triggered) another automation.

Each event carries one of five statuses:

Rerun an event

7. Click Rerun in the tab bar to run the automation again for the same candidate or record. In the Run Mode dialog, choose Normal mode (runs in the background; time blocks wait as usual) or Express mode (runs immediately and skips time blocks), then click Run.

A rerun uses the record's current data, not a frozen copy — so if the run failed because of bad data, fix the data first and the rerun picks it up. It also runs the automation's currently published version — not the version the original event ran on, and not an unpublished draft. So if a block was the problem, fix the block, publish, then Rerun the failed event; the employee doesn't need to resubmit the form. The rerun appears as a new event at the top of the list.

Rerun and Run do different things. Rerun replays a historical event against the record that event already ran for, so you can't point it somewhere else. Run, in the automation editor, targets any placement or task ID you supply — use that one when you want to try the automation on a different record. See 📄 How do I test an automation before publishing?

Confirm you found the cause

You're done when you can point to the exact block where the run stopped and the data it saw at that moment. From there the fix is one of three things: correct the source data and Rerun the event, adjust the condition or action in the editor, republish, and Rerun the failed event, or — if the trigger never fired at all — revisit the trigger setup in 📄 How do I build an automation?

Related

• 📄 Understanding automations

• 📄 How do I build an automation?

• 📄 How do I test an automation before publishing?

• 📄 Automation triggers and actions reference