Paid ads
Cutting CPA by 46% and Unlocking Growth for a $1M Clothing Brand
This DTC clothing brand, based in Singapore, has been in operation for eight years. They had built a steady presence in their market and achieved approximately $2.1 million in annual revenue, but when they attempted to scale to $3 million, everything fell apart.
They were spending $50k+ a month on Facebook ads, but results weren’t scaling with spend. Average order value (AOV) was ~$110, healthy for clothing but customer acquisition costs were eating all margin.
By June, the account had spent $51,765.16 with a CPA of $50.85 and a ROAS of just 2.31. Worse, frequency was 7.63, meaning the same people were seeing the same ads over and over again. In reality, their new customer CPA (based on Triple Whale) was closer to $59, which made scale nearly impossible.
That’s when we came in.
The Challenges
After auditing the account, three main bottlenecks stood out:
Bloated Ad Account Structure
38 campaigns running in a single market. Most are selling the same products to the same audiences. Overlap everywhere, learning phases never ending, and no clarity on what was actually working.
Lack of Creative Testing & Reach
Despite spending $286k year-to-date, they had only reached 2.3M people total. That’s a red flag. Why? Because Facebook is now a content-first platform. No new creatives = no new reach = no new customers. The agency reused the duplicate content for months, which caused reach to stagnate and frequency to skyrocket.
Tracking Vanity Metrics Instead of Business Metrics
The agency reported CPP and ROAS from Ads Manager every week but never looked at contribution margin, new customer revenue %, or actual profit. The brand had no system for tracking whether the spend was actually growing the business.
Our Process
Phase 1: Simplify the Structure
The first thing we did was pare down the account to the essentials. Complexity doesn’t scale; focus does.
One CBO campaign for the hero product.
Separate campaigns for top categories.
One small retargeting campaign with organic content.
That’s it.
When you have 38 campaigns, you’re not testing audiences, you’re cannibalizing them. Each campaign is targeting the same people, just split across different setups. That inflates frequency, confuses the algorithm, and spreads data too thin.
By consolidating into fewer, larger campaigns, we enabled the algorithm to learn more quickly, build data density, and consistently optimize for new customer conversions.
Phase 2: Build a Creative Testing System
Creative is the number one growth lever today. Yet this brand hadn’t introduced meaningful new ad concepts in months.
Here’s how we fixed it:
Make More of What Works
If a video or headline is pulling, we duplicate it with only one minor tweak. Same hook, new background. Same message, new angle. That way, we don’t waste time reinventing the wheel.
Make Better of What Works
Take a winning ad and level it up. Bigger headline, tighter hook, improved highlight, clearer CTA. The same idea, just sharpened.
Introduce New Concepts
Research competitors and customer desires. Then test new concepts. Not 20 random creatives, but a handful of well-thought-out ad creatives.
For this brand, clothing is both visually appealing and identity-driven. The most powerful ads helped customers see themselves in the clothes. Mirror try-on videos, “perfect outfit for [scenario]” reels, and lifestyle storytelling outperformed generic studio shots every time.
We also added a missing layer: representation. A 25-year-old blonde in an outfit only resonates with other 25-year-old blondes. We recreated winning ads featuring women of different ages and body types. That one adjustment opened up entirely new customer segments without needing new ad angles.
Phase 3: Track Business Metrics, Not Vanity Metrics
This was one of the most significant mindset shifts for the brand.
The old agency was reporting ROAS and CPP from Ads Manager. But those are vanity metrics. A 2.5x ROAS on paper doesn’t matter if new customer revenue is flat or margins are shrinking.
Instead, we aligned reporting to business metrics:
New Customer CPA
Contribution Margin
% of New vs Returning Revenue
Ecosystem ROAS (blended across channels)
Profitability by cohort
By focusing on these, we made decisions that drove profit growth, not just made the Ads Manager dashboard look good. For example, when CPA stayed high, instead of panicking, we focused on lifting AOV through upsells and bundles, which expanded margin room and allowed profitable scaling.
Results
June (Before)
Spend | Purchases | CPA | ROAS | Frequency |
|---|---|---|---|---|
$51,765.16 | 1,018 | $50.85 | 2.31 | 7.63 |
→ See FB Ad Performance in June
August (After)
Spend | Purchases | CPA | ROAS | Frequency |
|---|---|---|---|---|
$50,052.51 | 1,81 | $27.61 | 3.64 | 5.47 |
→ See FB Ad Performance in August
That’s a 46% drop in CPA
With the same spend, purchases increased by
78%
ROAS improved by
57%
Frequency dropped, reach expanded
7.4x
The brand went from being stuck at approximately $1M in revenue with razor-thin margins to finally unlocking profitable acquisition and preparing for a real scale path toward $4M+.
Key lessons for founders
Simplify before you scale.
Too many campaigns and hacks will kill efficiency.
Creative is your acquisition engine.
Without a testing system, you’re just recycling the same audience.
Track what actually matters.
ROAS is not profit. New customer revenue, contribution margin, and blended numbers tell the truth.
Build systems, not hero ads.
Rules, structures, and processes outlast any single creative.
Most agencies look at ads. We look at businesses.
That’s why we were able to cut CPA nearly in half, unlock volume, and set this clothing brand on track to scale profitably.
