Egnis.
Holding most of the media budget until the winning cut was obvious, and halving the cost of every acquisition after it.
Spend nothing in week one. Spend everything on what week one proved.
Egnis were amplifying every piece of creator content equally from the day it went live, which spends the same money against the best and the worst cut in a campaign. We held sixty per cent of the budget back and let the first week run organically to find out which was which.
Week one is cheap and unambiguous: three of the fourteen cuts were carrying the campaign, and one of the three was an outtake nobody had planned to publish. From week two everything went behind those three, in the formats the organic data said were working.
Half the cost per acquisition on the same creative and the same total budget.
Nothing about the creative changed and nothing about the budget changed; only the timing of the spend did. Cost per acquisition halved across the eight weeks, and the outtake that nobody had planned to publish became the single best-performing asset Egnis ran that year.
