Case Study: Paid Growth for BN3TH

unnamed (1)
Flywheel results
0 %

YoY Revenue Growth

Flywheel results


Flywheel results
0 x


Before Flywheel, BN3TH’s ad performance had begun to stagnate, and, because of how the Meta ad account was structured, had no way to scale.

We helped introduce two key frameworks for performance-led growth, smashed 2022’s targets, and blew BFCM out of the water.

Simon Leduc

Growth Marketing Manager

“Scott went completely sicko mode on this account.”


BN3TH specializes in premium men’s underwear and apparel. They launched their ecommerce store in 2015 and grew incrementally until the ecommerce rush of 2020/2021 helped them take off. They built an ads program to help them capitalize on it.

But their digital ads strategy had a problem: it wasn’t clear what was and wasn’t working. If you can’t isolate variables, then you can’t learn, and if you can’t learn, you can’t scale or respond strategically to seasonality or market downturns.

We partnered with BN3TH in February 2022. They had aggressive growth goals for the year and recognized their ads’ status quo wouldn’t take them to the scale they needed to achieve.



We helped BN3TH achieve its ambitious revenue target. While other ecommerce brands struggled in 2022, we accelerated.


Year over year:


Revenue growth


CAC decrease


Average stop rate increase

2s to 3s

Average watch time increase

Our joint efforts with BN3TH culminated in an unspeakably huge Q4, with a Black Friday campaign they’ll sing songs about for years to come—driving 50% YOY growth at a Marketing Efficiency Rate (MER) of over 10.


BN3TH has a great brand and a willingness to experiment. We worked with them to introduce two frameworks for performance-led growth.


First, we created a strategic, lumpy calendar to improve revenue throughout the year (you’ll see what we mean below.)

Second, we systematized their Meta account for optimal testing and scaling.

Seasonality: from Flat to Lumpy

The difference between 1 lump (2021) vs 3 lumps (2022)

BN3TH had a typical revenue curve for an ecommerce store: static revenue throughout the year with a hockey stick spike in the last two months. This is problematic for two reasons:

  • Product-based businesses live and die on cash flow. So ideally revenue is coming in at a predictable, regular cadence—as opposed to waiting for most of it to come in the final two months of the year.
  • Relying on two outsized months for sales is an inherently risky strategy. Especially with BFCM becoming increasingly competitive.

A Lumpy Calendar creates several moments throughout the year that give your audience the answer to the question: why now?

We worked with BN3TH to engineer demand around four A moments (in addition to a smattering of B and C moments):

  • Valentine’s Day
  • Father’s Day
  • End of Season Sale
  • BFCM.

A Lumpy Calendar generates steadier revenue throughout the year, but that’s not all. When it’s well done, a Lumpy Calendar compounds revenue with each successive moment as the customer base grows. We saw the same new customers from May buying again in July, and buying again at the end of the year.

Do not fade the Lumpy Calendar!

Systematized to Scale

You know the saying: “if it ain’t broke, don’t fix it?” That didn’t apply in this case.

When we first took over Meta, the account structure was technically performing, but it was murky and inefficient. Unfortunately, in this case, it needed a reset.

To drive long-lasting, repeatable growth on Meta, you need the trifecta of direct response:

  • Core audiences who resonate with your offer
  • Clear, compelling angles
  • Creative that captures the imagination

And there’s an order of operations you need to follow. You can’t start with creative if you don’t have your angles. And you can’t start with angles if you don’t know your audience. So, with just over 10 months left in the year, we set about building a foundation of core audiences.

Know thy audience. Know thy angles. Know thy creative levers.

1. Audience

We took a strict approach to audience testing: evaluating ad sets with different targeting, similar copy, and the same creative on a two week cadence.


In early 2022 we found lookalike audiences were struggling across accounts as they became saturated by automated bids. So we took an unconventional approach. We hypothesized on personas and validated with data.


We worked closely with BN3TH to create our first batch of personas to test, leveraging their SKU-specific sales, customer insights, and previous brand-building work.

Social proof ads to validate audiences. Not too fancy. But effective for gauging interest. 

2. Angles

With our core 5 audiences in place, we were ready to scale vertically by finding what messaging worked the best.

Horizontal scale is achieved by finding new audiences and marketing to them. Vertical scale means identifying better ways of communicating your offer to existing audiences.

We crafted a monumental document of copy and positioning—weighing in just shy of 10,000 words. 10k words might seem like overkill, but the difference between a losing angle and winning angle ended up being a 5x improvement in performance.

We took a 2-week testing cadence with our angle testing. Each angle occupied its own ad set in an ABO testing campaign to prevent Meta from Zucking around with our controlled test budgets.

Screenshot of our angle test tracker

3. Creative

Once we had our audiences and angles in hand, we could then invest more confidently in experimental creative, which is where you often get the most leverage in an ad account.



We worked with BN3TH’s extremely talented design team to iterate and improve creative throughout the year.


These incremental improvements helped us bring our average stop rates from <15% to >20%, and our average video watch times from 2s to 3s+.

There are 3 ways to approach creative: experimentally, incrementally, and modularly.

Experimental creative involves testing brand new creative formats.
Incremental creative takes existing creative formats that work and make minor, data-driven improvements.
Modular creative takes one image or video and dupes it out 2-15 times with small tweaks in order to identify winning combinations of variables.

Example of successful experimental creative

Putting it all together: Black Friday/Cyber Monday

Our Black Friday campaign was a beautiful culmination of a compounding, Lumpy Calendar and systematized testing.

Our customer base had grown all year, thanks to two A Moments. And months of testing audiences, angles, and creative put us in a strong position to succeed. So for Black Friday, we took proven creative formats, modularized them 3-8 times, and got them in front of proven audiences.

To modularize static images, duplicate the exact same piece of creative, but tweak 1 variable at a time. The idea is to isolate specific variables like text overlay, color, or featured SKU(s). The outcome is 3+ graphics that look basically the same but provide new insight on performance.

The result? 50% YOY growth with an MER >10.

Some selects from the batch of BFCM 2022 modular creative.

Let's Chat!

Reach out to start a conversation about how we can help grow your business online.