How Mejuri Discovered What Was Really Driving Sales

How Mejuri Discovered What Was Really Driving Sales

By partnering with Haus, Mejuri uncovered a key opportunity for growth on Meta — and boosted their iROAS by 57%.

“Haus has opened our eyes to the true incrementality of our marketing channels.”

Maddie Gaasenbeek
Marketing, Mejuri

The challenge

Meta was a key growth channel for Mejuri, but something didn’t add up.

Mejuri, a global fine jewelry brand redefining modern luxury for every day, partnered with Haus to understand the ROI of their paid media strategy. They were surprised to learn that while Meta ads drove revenue, they fell short of iROAS targets. Most of their revenue from Meta ads was coming from people who already knew and loved the brand.

This led Mejuri to wonder:

Could altering their Meta strategy help them efficiently reach new, incremental audiences?

The hypothesis

Mejuri suspected that upper-funnel campaigns might help convert new buyers.

Using Haus GeoLift, Mejuri designed an experiment to test whether upper-funnel Meta ads actually drive new customer revenue — something that typical platform reports struggle to measure.

They reallocated 25% of their Meta ads budget to upper-funnel campaigns and measured their incremental impact.

Create the experiment

Working alongside a Haus Measurement Strategist, Mejuri translated their hypothesis into an experiment that could isolate true lift.

Setup

With a few clicks, Mejuri began designing an experiment with Haus.

Experiment hypothesis

Upper-funnel Meta campaigns might help convert new buyers.

Geographical targeting

Location
United States
Australia
Canada
France
Germany
India
Japan
Mexico
United Kingdom

Granularity
Designated Market Area (DMA)
States

Target
All of United States
Include/exclude specific regions

Platform

Meta
Google
YouTube
TikTok
Programmatic
Apple
Amazon
Pinterest
LinkedIn
Snapchat
Linear TV
CTV
Out of Home
Streaming Audio
X (Twitter)
Linear radio
Direct Mail
Reddit
Microsoft
Affiliate
OTTY
Yelp
Multiple platforms

KPI
New Orders
New Revenue
Total Orders
Total Revenue

Platform

How would you like to measure its incremental impact?
Test lift with holdout
Adopt new marketing channel
Increase spend
Decrease spend
A/B test

Optimize

Haus lets you tailor your experiment design to your needs. Mejuri decided to balance power and precision by selecting a 3-week test with a 30% holdout.

Experiment optimization

New total orders
$X.XXX
Precision
Maximize precision 50/50 Split | 4 weeks
Minimize opportunity cost | 70/30 Split | 3 weeks

Best of both worlds
50/50 Split | 3 weeks

Check “Track the level of conversions after the experiment ends.”
This tracks delayed sales — like someone who clicked on an ad in Week 3 and purchased in Week 4.

Test length: – weeks (with 1 week PTW)

Apply experiment

Haus automatically applied the test and control groups to Mejuri’s Meta campaign.

experiment runs

The experiment ran for 3 weeks, with a 1 week post-treatment window.

Experiment results

Overall, Mejuri saw 1.5x better performance than their original iROAS goal.
By testing the impact of adding upper-funnel Meta ads into their media mix, Mejuri saw an 11% lift in incremental revenue and a more efficient iROAS.
After the two-week post-treatment window, results improved even further. They saw a 12.9% lift in incremental revenue and a 57% improvement in iROAS.

Mejuri upper-funnel Meta results

Tactic Meta
Test Type 2 cell w/ holdout
KPI New revenue
Test date Feb 22 – Mar 14, 2025

57% iROAS
12.9% Incremental sales

From Insight to Action

With this experiment, Mejuri was able to efficiently drive more revenue from new audiences.

Mejuri’s shift in media mix drove a 12.9% lift in incremental sales and a 57% improvement in iROAS.

By using holdout testing, Mejuri proved the impact was real. They now have a clear path to optimize media spend for greater efficiency and growth.

How confident are you in what’s actually driving your growth?

Just like Mejuri, a small, strategic shift might have a big impact. You’ll never know unless you test.

The Laws of Incrementality

Whether you’re new to incrementality or a testing veteran, The Laws of Incrementality apply no matter your measurement stack, industry, or job family.