Decide what the ad platform may change automatically

Inventory the active automation

Inspect which recommendation categories are currently opted in, who enabled them and when. Google documents auto-apply settings and history for reviewing these choices. Preserve a dated inventory before changing anything.

Include other automation sources that can affect the same campaigns, such as approved rules or external management tools. The goal is to know which systems can act and under whose authority. Avoid assuming that every unexplained change came from the platform’s recommendation feature.

Evaluate categories one by one

For each category, write the action it can take and the business condition under which that action is desirable. Consider effects on audience, messaging, bidding and measurement. A bundle name can hide changes with very different consequences.

Check the exact current behavior and permissions. Identify actions requiring prior human review because of spending limits, claims, brand restrictions or contractual commitments. The platform’s ability to make a change does not establish that the business has authorized it.

Set review and exception rules

Name an owner, a review cadence and a small set of signals that trigger investigation. Track actual changes and their effect on the intended outcome rather than counting how many recommendations were applied. An increase in a platform score is insufficient evidence of business value.

Specify how to pause future automation when necessary and who approves any reversal of previous changes. Turning off a category may stop future actions without undoing what already happened. Verify the supported behavior before promising a complete rollback.

Keep a readable change record

For material changes, retain the before-and-after setting, source of the action, affected campaigns and reason for keeping or reversing it. Use this record when interpreting performance shifts and preparing account handovers. Avoid relying on memory about who probably made the edit.

Periodically reassess whether the original authorization still fits the business. A new product, legal restriction or account owner can change the acceptable scope. Review the agreement with the responsible person rather than letting inherited settings become permanent policy by default.

Make the reasoning concrete.

A hypothetical agency inherits an account with several auto-apply categories enabled. A new campaign setting appears, and the team initially blames a junior operator. The history can identify whether the action came from an approved automation.

Verify the history before assigning blame. Keep, narrow or disable automation through an authorized, category-specific decision.

Take the questions into your next brief.

Use these prompts in your own notes. Nothing is submitted from this page.

1. Inventory

   Which systems and categories can act?

2. Authority

   Who approved each action scope?

3. Fit

   When does this category help the business?

4. Audit

   Which changes and outcomes will be reviewed?

5. Rollback

   What stops future actions and what reverses prior ones?

Watch for the wrong conclusion.

• Opting into bundles without reviewing individual actions.
• Treating platform recommendations as business authorization.
• Assuming disabling automation undoes previous changes.
• Attributing unexplained edits to a person without history evidence.

Sources & scope

Frameworks, worksheets and hypothetical examples are original editorial synthesis. Linked primary sources support only the stated concepts. Source checking is not an independent audit of effectiveness. No named human author or reviewer is claimed.

• Google Ads Help: Manage auto-apply recommendations →Primary platform documentation · Supports: Documents selected recommendation categories, opt-in history and management controls. Platform automation availability does not supply the business owner’s authorization.