Campaign QA is the set of checks that should happen before a campaign launches and every day while it runs. Done by hand, it is tedious, inconsistent and the first thing skipped when a team is busy. Done by automation, it is reliable, fast and almost entirely low-risk — because QA mostly reads rather than changes.
Pre-launch checks to automate
- Final URLs resolve and land on the intended page, with tracking parameters intact.
- Conversion tracking fires on the destination, and the right conversion actions are set as primary.
- Budget and schedule match the plan — no extra zero, no wrong flight dates.
- Targeting is right — geography, language, audiences and exclusions.
- Naming follows the convention, so reporting and later automation can identify the campaign. See campaign standardisation.
Daily checks to automate
- Pacing against the monthly plan. See PPC budget automation and pacing.
- Disapproved or limited ads, and policy flags.
- Budget-limited campaigns that are losing impression share they could afford.
- Spend with no conversions beyond a threshold.
- Broken landing pages that appeared since launch.
- Tracking anomalies — a sudden drop to zero conversions usually means a broken tag, not a bad day.
The output is a short daily digest of exceptions, not a dashboard. The people who need to act should see only what needs acting on.
How approvals should work
Approvals are where speed and control collide. The pattern that works is tiered: each type of change sits in one tier — recommend, act with approval or act and log — and anything not explicitly listed defaults to the most restrictive. An approval request should show a clear before-and-after difference, not a description. And an approval that expires is a rejection: silence is never consent. See approval thresholds and the three action tiers.
What a real audit trail looks like
As platforms and agents keep more persistent context about your business, the audit trail matters more, not less. A useful trail records, for every change: the specific condition that triggered it, the guardrail check, the change itself with the state before it, who or what approved it, and the measured result afterwards. Native change histories in the ad platforms show what changed; they rarely show why. See the AI marketing audit trail and trigger, action, impact.
Which platforms support automated QA and approvals?
The ad platforms themselves offer partial help — change history, policy status, automated rules and alerts — but little in the way of structured, cross-platform approval workflow. Third-party optimisation suites add alerts and some approval steps. Many teams get the most complete result by building QA and approvals inside the tools they already use, so the checks reflect their own plan, naming and thresholds. The campaign launch kit and automation governance build guides show how.