Inspect an automation run
To see what an automation did, open its recent runs, pick one, and read each step's status and, for a failed step, its error. A run can look successful and still have done nothing, such as sending zero emails, so read the skip reasons too. Fix the cause, check it with a dry run of the step, then replay the run.
To see what an automation did, open its recent runs, pick one, and read each step's status and, for a failed step, its error. A run can look successful and still have done nothing, such as sending zero emails, so read the skip reasons too. Fix the cause, check it with a dry run of the step, then replay the run.
Requirements
A role that can view and edit automations.
Steps
- Open the automation's recent runs. Each run shows its status. If there are no runs at all, the automation never started; see My automation did not run.
- Open a run. Every step shows whether it ran, was skipped, or failed. A failed step shows its error; a skipped step shows why.
- Read the skips, not just the failures. A send-email step that found no recipients, or whose body came out empty, is skipped rather than failed, so the run looks fine while nothing was sent. A staffing firm's welcome automation ran green for a week; its runs showed every email skipped because the candidate email field was empty on entries created from the site.
- Reproduce the failing step safely. Dry-run that one step: it runs with nothing saved and nothing sent.
- Fix the cause, in the entry data or the automation.
- Replay the run, or start the automation by hand, and inspect the new run.
Common findings
- A temporary error, such as a timeout, a network failure, or the other service being briefly unavailable. Add retries to the step, where the step allows them; see Steps and actions overview.
- A run stuck running or waiting in an automation with branches. Two branches that rejoin a shared step leave it waiting for a branch that never comes. Put the condition on the step instead of branching.
- A switched-off automation. Replaying a real run or starting it by hand is refused while it is off. Dry runs and test runs still work, so you can fix it before switching it back on.
Notes and limitations
- AI steps cost money when replayed. An agent will not replay or hand-start a run that contains an AI step; it checks with a dry run and asks you to run it. See Add an AI step to an automation.
- Scheduled automations can have excluded dates; a run missing on a certain day may be an exclusion.
- An agent can do this whole diagnosis for you; ask it to follow its recipe for a failed or silent automation. See Workflow recipes for agents.
Related
For what the steps are, see Steps and actions overview; for what starts a run, Automation triggers.