A rapidly growing e-commerce brand had a handful of ad creatives posting exceptional returns while much of its budget bled out on underperformers. We built a high-velocity testing and budget-allocation system that concentrated spend on proven winners and kept discovering new ones.
The challenge
Client A, a rapidly growing e-commerce brand, experienced significant volatility in ad performance. A few key creatives showed exceptional Purchase ROAS of up to 31.18, but a large portion of the budget was wasted on underperforming ads with low returns, making it inefficient and unpredictable to scale.
Ruthless pruning
We implemented a strict A/B/C testing framework to quickly identify and immediately pause any creative that failed to hold a minimum profitable ROAS threshold, which we set at 3.0. This instantly reallocated budget away from weak performers and toward proven winners.
Creative replication
We used data from the highest-performing ads to deconstruct their core success elements such as hook, offer, and format. New creatives were built to replicate and test these variables across different audiences, ensuring a sustained pipeline of new profitable winners.
Targeted scaling
We shifted the bidding strategy to focus entirely on Value Optimization, letting the platform aggressively bid on users likely to generate the highest return. This meant scaling the budget did not dilute the performance of the top ad sets.
The results
The best-performing ad achieved an exceptional 31.18x Purchase ROAS, and we identified 7 winning creatives that held an average ROAS of 9.44x or higher. By redirecting budget from underperforming ads, we delivered a 45% improvement in overall campaign efficiency.
