Testing Your Automations

Test before saving

The recommended approach is to test a workflow with execute_steps before saving it as an automation. execute_steps uses the exact same step format as saved automations but does not persist the results as a run record. This lets you iterate quickly without cluttering your Workflow History.

Testing a saved automation

After saving, use test_automation to verify the automation runs correctly. test_automation executes the automation and returns step-by-step results. If the automation expects external input, pass test data in the initial_inputs parameter to simulate what would happen when real data arrives via webhook or API.

Running manually from the UI

Run any automation manually from the My Automations page by clicking the Run button. This is the simplest way to verify a new automation works against your real production data before enabling its schedule.

Debugging with Workflow History

Every automation run is logged in Workflow History. Clicking on a run shows whether each step succeeded or failed, the full output of each step, error messages for failed steps, and timing. This is the primary debugging tool for production automations.

Test scenarios to cover

• Happy path — the main scenario with representative real data
• Empty data — what happens when an API returns zero results
• Webhook input — test with sample data matching your expected payload
• Approval flow — if the automation has a human approval step, verify both the approve and reject paths work