Google Ads Audit Checklist: What to Review Before You Optimize
An account audit only holds up if it starts in the right order. Here is the sequence that avoids optimizing on top of broken measurement.
Most Google Ads audits start with the wrong question. They start with “what should we change?” when the first question should be “can we trust what we’re measuring?” Optimizing bids, budgets or keywords on top of broken measurement doesn’t just waste effort — it actively makes the account worse, because every automated bidding decision downstream is trained on the same bad signal.
This is the order we’d review an account in, and roughly why each step has to happen before the next one is worth doing.
1. Conversion tracking, before anything else
Check whether there is at least one conversion action that is both enabled and actually recording conversions. An account can look “set up” — the tag fires, the action exists — and still not be counted toward bidding if it’s disabled or misconfigured. Then check which actions are marked primary. It’s common to find five or six primary conversions on an account: the real purchase event, plus add-to-cart, page views, and a newsletter signup, all weighted equally. Smart Bidding doesn’t know which of those matters to the business — it just optimizes toward whatever is marked primary. Micro-conversions mixed in with revenue events dilute the signal the algorithm is actually chasing.
Also check for duplicate primary conversions — the same purchase counted twice because both a native Google Ads tag and a GA4 import are active and both marked primary. This is one of the more expensive, least visible problems we see: reported conversion volume looks healthy, ROAS looks inflated, and nothing in the interface flags it as a duplicate.
2. Search terms, before touching keywords
Pull the search terms report before you touch match types or add negative keywords. Look specifically for terms with meaningful spend and zero conversions over a long enough window that the zero isn’t noise. A handful of clicks proves nothing; a few hundred dollars of spend with no conversions across weeks is a pattern. This is usually a faster, more defensible source of savings than re-tuning bids, because it’s a direct waste number, not an estimate.
3. Budget allocation
Look at where spend is concentrated relative to where performance is concentrated. It’s common for one or two campaigns to absorb the majority of the account’s budget while running meaningfully below the account’s target ROAS or CPA — often because the budget was set based on history, not current performance. Separately, check for campaigns that have spent a large multiple of target CPA with zero conversions in the period; that’s a different problem than concentration, and it deserves a different response.
4. Performance Max asset completeness
If the account runs Performance Max, check whether each asset group has the number of headlines, descriptions, images and logos Google recommends as a minimum. An incomplete asset group doesn’t just look worse — it limits which ad formats and placements the campaign is eligible for, which caps performance before you’ve touched a single bid setting.
5. Account hygiene
Last, not because it doesn’t matter, but because it’s the layer least likely to be actively losing money right now: campaigns marked enabled that aren’t actually serving, ad groups with a single ad and nothing to test against, disapproved or policy-limited ads sitting unnoticed. These are worth fixing, but fixing them before tracking and search-term waste is solving the quiet problem before the loud one.
Why the order matters
Every step after tracking inherits whatever error tracking introduced. A budget reallocation based on ROAS numbers that are 40% inflated by duplicate conversions isn’t a small mistake — it moves real money toward the wrong campaigns with confidence. That’s the reasoning behind how Ads Audit orders its own rule categories: tracking first, because every other recommendation depends on it being right.