The Challenge

RefrigiWear entered the holiday season with a 3.30 ROAS from the prior year’s November — a solid baseline, but not good enough for their growth targets. The core problem: branded and non-branded search traffic were competing in the same campaigns, which meant bottom-funnel branded demand was being diluted by top-funnel prospecting spend. CPAs were rising and performance had hit a ceiling.

What We Did

The fix was structural: separate branded search campaigns from non-branded ones entirely.

This is a classic mistake in paid search. When branded and non-branded queries compete in the same campaign, branded keywords often win the budget because they have higher Quality Scores and lower CPCs. This looks efficient in the data, but it means you’re capturing demand that would have converted anyway — people who were already going to search for RefrigiWear — while underfunding the prospecting campaigns that actually grow the customer base.

We isolated branded campaigns to protect bottom-funnel conversion efficiency while giving non-branded prospecting campaigns their own dedicated budget and optimization targets. This allowed both to perform at their best without cannibalizing each other.

On Meta, we ran awareness and consideration campaigns timed to the November–December cold weather and gifting window, driving top-of-funnel demand that the search campaigns could then capture. The combination — with proper campaign separation — drove a 57% ROAS improvement over two months.

The Results

• 57% ROAS growth — from 3.30 to a significantly higher blended return in just two months
• 30% revenue growth — YoY improvement during the critical holiday window
• 20% BFCM revenue lift — ~$300K in tracked incremental holiday revenue

The Takeaway

If branded and non-branded search are in the same campaign, your data is lying to you. Separation is the single highest-leverage structural change most paid search accounts can make — and it costs nothing except the time to set it up correctly.