Google Ads audit · Athletic apparel
The account was working. The structure wasn’t finished.
An agency asked Channelshift to audit a client account that was meeting ROAS benchmarks and generating conversions. The review found six structural gaps and produced a seven-step rebuild plan.
The brief
Look beneath an acceptable headline.
The account was not presented as a failure. It was hitting its stated ROAS benchmarks, producing conversions, and keeping the end client satisfied. That made the audit a question of structure: did the blended result describe acquisition accurately, and was the account positioned to capture more demand without simply increasing bids?
The review found branded campaigns supporting the blended CPA and concealing weaker non-brand economics. It also found materially different conditions inside Performance Max: one campaign had strong ROAS but only 32% impression share, while another remained active at a $185 CPA.
Evidence source: the disclosed account figures were sourced from platform data. The public source does not identify the client or agency.
The audit
Six structural findings.
Brand obscured acquisition cost
Branded campaigns supported a blended $33 CPA, making the underlying non-brand acquisition picture harder to see.
Branded impression share stopped at 72%
The branded campaign was not capturing all available branded demand.
PMAX performance was uneven
The strongest PMAX campaign was limited to 32% impression share despite strong ROAS, while another remained active at a $185 CPA.
Every PMAX campaign used one asset group
The structure limited the account’s ability to organize and test distinct product or positioning inputs.
Search and Shopping coverage was incomplete
The account had no standard Search, Dynamic Search Ads, or standard Shopping campaign alongside PMAX.
Auto-applied recommendations were enabled
Google could apply platform recommendations automatically rather than leaving each account change to an explicit operator decision.
The deliverable
A seven-step proposed rebuild.
- Turn off auto-applied Google recommendations.
- Add asset-group diversity to each PMAX campaign and test signals-only groups.
- Move the branded campaign to Target Impression Share at 100%.
- Extract converting search terms from PMAX and build exact-match Search campaigns.
- Launch a Dynamic Search campaign to surface missed long-tail demand.
- Launch standard Shopping at a target CPA below the best PMAX campaign to capture remaining demand.
- Add PMAX Insights and Market Forces scripts for ongoing structural monitoring.
The plan emphasized account control and coverage rather than a larger budget. It was intended to make brand and non-brand performance easier to interpret, expose missed demand, and create a testable structure.
Evidence limits
What this case does—and does not—show.
This was an audit and planning engagement. The record documents the account’s observed state, six findings, and the seven-step proposal. It does not report whether the client implemented the plan.
No post-change CPA, CPC, ROAS, revenue, conversion lift, or budget change is available. Lower CPCs, additional scale, and better efficiency were expected opportunities identified by the plan—not measured outcomes. They should not be read as results delivered after implementation.
The client and partner agency are unnamed. The public record does not specify the client’s sales volume, media spend, audit duration, currency, or a formal confidentiality agreement. The business is labeled athletic apparel; describing it more narrowly would require additional client-approved evidence.
Disclosure: account figures above are presented as recorded in the original case-study source. No implementation result has been added or inferred.