GeoLift
Run experiments to understand the business impact of your marketing — from digital ads to OOH campaigns —across all your sales channels.
Experiments = Incrementality
Experimentation is the universally agreed upon way to answer the most important question in marketing:
What would have happened if we hadn't run this campaign?
Without testing, you're just guessing—and geo-experiments give you the clearest answer by comparing similar markets where you did and didn't advertise.
Types of GeoLift Experiments
Standard Haus Geo Tests
Random assignment of test and control groups across the country to give you strongest results.
Best suited for: National digital (Meta, Google, TikTok, Pinterest) or offline campaigns in which you can request to turn on and off media in certain regions.
Fixed Geo Tests
You select the test group and Haus uses advanced modeling techniques to create a Haus Holdout.
Best suited for: Regionalized advertising (OOH campaigns, regional linear TV, events, retail, direct mail), in which you have a pre-selected treatment group.
Navigate growth with a reliable compass
Less accurate
More accurate
Attribution
Powered by Clicks
- Limited by privacy changes
- Can’t measure offline
- Undervalues upper funnel
Traditional MMM
Powered by Correlation
- Depends on outdated data
- Lacks frequency and granularity
- Time and cost intensive
Incrementality
Powered by Causality
- Test/Control methodology
- Based on actual sales data
- Fast and actionable
Use GeoLift for:
Incrementality
Holdout testing to understand how marketing treatments drive sales.
Read more
Optimal spend level
Plot your spend on a diminishing marginal return curve to find the right investment levels.
Read more
Amazon + Retail
Measure across all sales channels including DTC, Amazon, and retail.
Read more
Media mix optimization
Compare the performance of two marketing treatments against one another.
Read more
New channel launches
Quantify the impact of launching into a new marketing channel.
Read more
Holidays and promotions
Identify the incrementality of discounts & promotions.
Read more
From question to experiment in minutes
Configure robust regional experiments on-demand with the statistical tools and controls to balance speed and precision.
How it works
An automated end-to-end solution, built by scientists, made for marketers.
You
1. Hypothesis
State your business question
Haus
2. Experiment Power
Configure test for your desired statistical power
3. Region Selection
Stratified sampling to split regions into test and control
4. Test Implementation
Execute campaign in the target regions with automatically implemented geo-holdouts
5. Statistical modeling
Increase precision with advanced synthetic control methods
6. Analysis & Insights
Translate results into actionable insights
We knew our Ritual customer was on TikTok, but there was uncertainty around whether as a new channel, it could be an evergreen part of our mix. We are always trying to understand the incremental effects of our advertising, and with Haus, we were able to optimize our TikTok strategy.
There has been a lot of industry chatter about whether PMAX campaigns overdeliver to brand terms and users who are not incremental. We used Haus to test and validate our assumptions, and the business impact was significant.
