Auto-Approve Rules
How to configure rules that automatically approve routine, low-risk changes in the Safety Gate.
Auto-Approve Rules
Not every change needs a human reviewer. When you sync employee data from Workday and 200 new hires arrive on the first Monday of the month, you probably do not need to manually approve each one. Auto-approve rules let you define conditions under which changes are approved automatically, so your team can focus review time on the changes that actually matter.
When to Use Auto-Approve
Auto-approve rules are appropriate for changes that are:
- High volume and low risk -- New hire records, routine department transfers, standard GL postings
- Predictable in pattern -- The change type and affected fields follow a consistent pattern
- Low financial impact -- Changes that do not affect compensation, revenue, or other sensitive figures
Do not auto-approve compensation changes, GL reclassifications, executive-level org changes, or any data that feeds directly into board-level reporting. These should always pass through human review.
Creating a Rule
Open Auto-Approve settings
Navigate to Safety Gate > Auto-Approve Rules for the pipeline you want to configure.
Click New Rule
Each rule needs a descriptive name (e.g., "Auto-approve new hires below VP level").
Define conditions
Set the conditions that must ALL be true for the rule to fire:
- Change type -- ADD, UPDATE, or DELETE (select one or more)
- Field conditions -- e.g.,
job_level != "VP"andjob_level != "C-Suite" - Risk threshold -- e.g., only auto-approve if risk is LOW
- Connector -- optionally limit the rule to a specific source connector
Set priority
Rules are evaluated in priority order (lowest number first). When multiple rules match a change, the highest-priority rule wins.
Enable the rule
Toggle the rule to Active. It takes effect on the next sync.
Rule Conditions in Detail
Conditions use a simple expression syntax:
field_name operator value
Supported operators:
| Operator | Meaning | Example |
|---|---|---|
| == | Equals | department == "Engineering" |
| != | Not equals | job_level != "C-Suite" |
| > | Greater than | amount > 0 |
| < | Less than | amount < 10000 |
| in | Value in list | change_type in ["ADD", "UPDATE"] |
| contains | String contains | account_name contains "Revenue" |
You can combine multiple conditions. All conditions must be true for the rule to match (logical AND).
Rule Priority
When multiple rules could apply to the same change, priority determines which one fires. Lower numbers mean higher priority.
For example:
| Priority | Rule | Action | |---|---|---| | 1 | Compensation changes: never auto-approve | Block auto-approve | | 2 | New hires below VP: auto-approve if LOW risk | Auto-approve | | 3 | Department transfers: auto-approve if LOW risk | Auto-approve |
In this setup, a new hire record that includes a compensation field change is caught by rule 1 and routed to manual review, even though rule 2 would otherwise auto-approve it.
You can create blocking rules (priority 1, condition matches sensitive changes, action = "require manual review") to ensure certain changes always require human approval regardless of other rules.
Audit Trail
Every auto-approved change is logged with full traceability:
- Which rule triggered the approval
- When the auto-approval occurred
- The full change detail (before/after values, risk score)
- Who created the rule that performed the auto-approval
Auto-approved changes appear in the Safety Gate history with an "Auto-approved" badge so they are visually distinct from manually approved changes. You can filter the history to show only auto-approved changes for auditing.
Monitoring and Adjusting
Review your auto-approve rules quarterly. Check:
- Hit rate -- How many changes did each rule auto-approve? Rules that never fire should be removed.
- False positives -- Did any auto-approved change turn out to be problematic? If so, tighten the conditions.
- Coverage -- Are there routine changes that your team still reviews manually? Consider adding a new rule.
The Auto-Approve Analytics panel on the Safety Gate dashboard shows rule hit counts, trend over time, and the percentage of total changes handled by auto-approval.