Services · Analytics
Testing and incrementality
A/B and incrementality experiments that prove what works before you scale spend.
Donut Analytics runs A/B tests and incrementality experiments, such as brand lift studies, to find what works before you scale spend. An incrementality test compares people who saw your ads with a control group who didn't, so you see what the ads added, not just what happened.
What’s included
What we test
-
A/B tests
Two versions of an ad or a page, head to head, before the budget goes behind one.
-
Brand lift studies
People who saw the ad against a control group who were eligible but didn't.
-
Creative comparisons
Not an experiment, but a useful read: how each cut plays to the end against the others on the same screens.
-
Tests and models together
A model is an estimate, not an experiment. A test gives a controlled read on one question.
Proof
One brand lift study, by the numbers
Real client work, kept anonymous. Each number links to how it was measured.
- YouTube Brand Lift +63% awareness lift among 55–64s for our lead YouTube spot, at $0.23 per lifted user. 6.73% of people who saw the spot vs 4.13% of a control group. Overall awareness rose 7%. Not every age group or device showed lift. See how it was measured
- YouTube Brand Lift $0.23 per lifted user among 55–64s, against $1.02 across the whole study, blended. Lower is better. Same study as the awareness lift. See how it was measured
- YouTube Brand Lift ~20% more aware on TV screens: 9.01% of people who saw the spot vs 7.54% of the control group. From the same brand lift study, Dec 21–24, 2025. See how it was measured
FAQ
Questions about testing and incrementality
- What is incrementality testing?
- It measures what your ads added by comparing a group that saw them with a control group that didn't. The difference between the two is the lift.
- What is the difference between A/B testing and incrementality testing?
- An A/B test compares two versions against each other. An incrementality test compares ads against no ads, to see whether the ads moved anything at all.
- What did a brand lift study show for a client?
- For a luxury resort launch, people aged 55–64 who saw the lead YouTube spot were 63% more aware than a control group who didn't (6.73% vs 4.13%). Overall awareness rose 7%. Not every age group or device showed lift.
Related services
- Analytics
Marketing mix modeling
Media mix models on Google's Meridian that measure ads against all sales.
Learn more - Marketing
YouTube and connected TV
YouTube, Demand Gen and streaming TV, measured with lift studies and models.
Learn more - Analytics
Measurement and attribution
Clean tracking and attribution in GA4, GTM and BigQuery, tied to revenue.
Learn more
Guides: Incrementality vs. attribution: what's the difference?How to measure streaming TV ads FAQ
Test it before you scale it.
Tell us what you’re working on. We’ll build a measurement-first plan for your brand. Email hello@donutanalytics.com or call (407) 394-9720.