The Weakest Campaign in a Strong Google Ads Account: What the Average Hid
By Joseph Maina · · 3 min read
An account reporting 15.5x return on ad spend sounds like an easy success story. But the average can hide a very different result inside one campaign. This article revisits the published Google Ads launch case study for an anonymised Kenyan beauty retailer, focusing on its weakest campaign. It is a closer analysis of the existing report, not a new client result or a new measurement period.
Start with the scope of the original report
The published case study covers 1 August to 8 October 2026. It records KES 1,942,243 in platform-tracked revenue against KES 125,422 of spend, with 339 tracked orders and 15.5x overall return. The report converts Google’s figures at KES 129.89 per US dollar.
Those are attributed reporting figures, not audited profit or proof that every order was caused by advertising. The original report also notes gaps in its weekly log, so the account total and logged breakdowns should not be forced into a perfect reconciliation. Keep the original scope visible when quoting a smaller part of the account.
Sources and further reading
The consultation campaign had different economics
The local skin consultation search campaign recorded KES 10,257 of spend, KES 81,265 of tracked revenue and 13 tracked orders. Its reported return was 7.9x and cost per tracked order KES 789. General product search reported 18.0x and KES 326 per tracked order.
The useful observation is the difference inside the same account. An overall figure can look strong while one service or offer acquires an attributed result at a much higher cost. That difference deserves investigation even when the campaign itself reports revenue above spend.
| Published campaign or ad group | Tracked ROAS | Cost per tracked order |
|---|---|---|
| General product search | 18.0x | KES 326 |
| Shopping, all products | 14.9x | KES 356 |
| Local skin consultation search | 7.9x | KES 789 |
The report identifies the gap, not its complete cause
The available figures do not establish whether the weaker result came from the search traffic, landing page, offer, appointment process or another factor. Nor do they establish the margin on a consultation-related order. A convincing explanation would need more evidence than a return column.
Before recommending a change, I would inspect the conversion definition, search intent, landing experience and business-side outcome. Those are proposed next checks, not actions this article claims were performed. Distinguishing the observation from the explanation keeps the case study useful.
A lower return is a question to investigate
A consultation may serve a different commercial purpose from an immediate product sale. To evaluate that possibility, the business would need reliable downstream records and an agreed comparison period. It would be speculative to claim future value from those 13 tracked orders without that evidence.
The practical lesson is to review material campaigns individually and connect their reported outcomes to business economics. Neither keeping every campaign because the account average is strong nor pausing the weakest line automatically is a complete decision process. The next step should answer a defined uncertainty.
Ask what your own account average leaves out
In a campaign review, ask to see the important services, product groups or campaign lines alongside the total. Bring the commercial outcomes you can verify. I can scope a marketing review that separates the reporting facts, their limits and the next questions worth testing.
Sources and further reading