Help Center

Docs

Runs & Analytics

Every run, recorded: run history, audit cards, goal conversion, re-enrollment, and which version each run used.

The health header

A quick read on whether an automation is doing its job.

Open any automation and the header summarizes its recent health: an overall status of Healthy, Degraded, or Idle, plus its last run, next scheduled run, runs in the last 7 days, and its success rate. A degraded automation is one whose recent runs have been failing.

Run history

One row per run, with everything needed to trace what happened.

The run history table lists every time the automation fired. Each row shows the record it ran on, which trigger fired it, when it started, how long it took, and its status. A fresh automation reads No runs yet until its trigger first fires.

FieldDescription
RecordThe record this run acted on, linked so you can open it.
TriggerWhich of the automation’s triggers started this run.
StartedWhen the run began.
DurationHow long the run took end to end, including durable waits.
StatusCompleted, failed, or still in progress. A run that paused on a wait or an approval shows as in progress until it resumes.

The run timeline and audit cards

Open a run to see every step it took.

Selecting a run opens its timeline as a drawer over the builder. Each step that ran appears as an audit card recording what changed, the before-and-after of an updated field, the task that was created, the email that was sent, so you can see exactly what the automation did, not just that it ran. Failed steps are flagged so you can spot the cause.

For AI Agent steps the card is the same card: what changed, whether the step succeeded, when it ran, and the note the agent recorded about why it acted. How the action was decided is not on the card, because that decision lives in the approvals queue.

Goal exit and conversion rate

Measure whether the automation is moving records toward the outcome you wanted.

If you set a goal on the automation, a record that meets it exits the flow early, because there is no point continuing once it has succeeded. The Hub turns those exits into a conversion rate: the share of enrolled records that hit the goal, shown as a percentage with the underlying counts, for example Conversion: 42% (89 / 212).

Re-enrollment in the history

Why a single record can appear more than once.

When an automation allows re-enrollment, the same record can run through it multiple times, so it appears once per run in the history. With the default Once ever setting each record runs a single time. Re-enrollment is configured in the automation’s settings; see Building an Automation.

Version drift

Each run remembers the version it enrolled on.

Editing and republishing a live automation creates a new version. Records already mid-flow keep running on the version they enrolled on, so a change you make today never alters a run that started yesterday. That is why an older run in the history may not match the flow you currently see in the builder: it ran an earlier version. New enrollments use the latest published version.