We cut a US subscription fragrance brand's spend on zero-conversion ads by 76% while putting 80% more creative into test
76%
decrease in spend on zero-conversion ads
30%
decrease in time spent on manual optimisation and testing
1.8x
more split tests
running at once
"They run testing structures on our account that we could never have staffed in-house — several strategies live against each other at once, and budget moved onto the winners the same day. We get the results without the daily management, and it shows up in what our Meta spend returns."
Challenge
The business runs on recurring revenue. A member pays a flat fee every month, so a new customer only pays off if the cost to acquire them stays well under what they go on to spend. That makes paid social the growth lever and the acquisition cost the number everything is judged on.
It came to us as a fast-growing subscription business with one objective: scale acquisition on Meta as cost-effectively as possible, and stop paying for ads that were bringing few or no conversions. The in-house team had the budget to grow with. What it did not have was the hours to work the account every day.
A catalogue wide enough that the creative angle, not the individual product, decides what a campaign returns.
Solution
Split (A/B) testing is how this brand finds the right combinations of ad content, targeting, placement and bidding. With a significant budget at stake, it needed far more of those tests running than its own team could set up and read — without adding headcount to campaign management.
It hired us to run paid acquisition on Meta. We buy against the metrics that matter to the business — cost per click, cost per acquisition, campaign budget, time-of-day performance and return on ad spend — and we monitor ad performance daily rather than at a monthly review.
More tests, running at once
Over the first three months we increased the number of split-test campaigns live on the account, segmenting ad sets by audience, placement and creative angle so each segment carries its own play. One ad strategy is always being tested against another, instead of one idea being trialled at a time.
Budget and bids moved daily
We raise and lower budget, manage bid actions, and start, pause and restart campaigns on what the results actually show — the same day they show it, not at the end of the month.
Zero-conversion ads cut early
Ads that spend without converting get caught in the daily review and stopped before they take a meaningful share of the budget. What they were absorbing goes back into the tests that are working.
Results
Three months of buying on Meta scaled the programme while cutting both work hours and wasted ad spend:
- 76% less spend on ads that brought no conversions
- 1.8x more split tests running at any one time — 80% more creative angles read per quarter
- 30% less of the in-house team's time spent on manual optimisation and testing
