Marketing
Incrementality
Also called
- Lift testing
- Holdout
Every attribution model is correlational. Incrementality testing is the causal version: hold a channel back from a randomized group, or turn it off in matched geographies, and measure the difference in outcomes.
The results are frequently uncomfortable. Branded paid search and retargeting are the two channels most often found to be substantially less incremental than their reported ROAS, because they intercept demand that would have converted anyway.
It costs real money to run — you are deliberately withholding marketing from part of your audience — which is why it is worth doing on your largest line items and not on everything.
One incrementality test on your biggest paid channel can be worth more than a year of dashboard refinement, because it tells you whether the number you have been optimizing is real.
Commonly misunderstood
What people get wrong
The claim
“Our ROAS is 6x, so the channel is working.”
What is actually true
ROAS says the channel was present when revenue happened. Incrementality says whether the revenue needed it. They can differ by most of the number.
Where this comes up
Services where it matters
Related terms
Next step
Working through a incrementality decision?
Tell us the situation. We will give you the tradeoffs as we see them, including when the answer is that you do not need what you are being sold.
No pitch deck. A 30-minute conversation about what you are trying to achieve.