An experimental method that measures whether a marketing activity caused an outcome that would not have happened otherwise. Incrementality testing isolates causal impact by comparing a group exposed to the marketing with a holdout group that was not.
Incrementality testing answers the hardest question in marketing measurement: did this marketing activity cause conversions that would not have happened without it? Attribution tells you which touchpoints a customer interacted with before converting. Incrementality tells you whether those touchpoints made a difference.
The method is experimental. You split an audience into 2 groups: a test group that sees the marketing (an ad, an email, a campaign) and a holdout group that does not. Both groups are measured on the same outcome. The difference in conversion rates between the 2 groups is the incremental lift, the portion of outcomes that the marketing caused rather than captured.
Correlation is not causation
Most marketing measurement tracks correlation, not causation. A customer saw an ad and then converted. But would they have converted anyway? Retargeting ads are the classic example: they follow people who already visited your site, many of whom would have purchased without the ad. Attribution credits the retargeting. Incrementality testing reveals whether it moved the needle.
This distinction has direct budget implications. If a channel shows strong attribution performance but low incrementality, you are spending money to take credit for conversions that would have happened organically. Redirecting that spend to channels with proven incremental impact improves overall efficiency.
Test design discipline
The first mistake is running incrementality tests without sufficient sample size or duration. Small audiences and short test windows produce noisy results that look like insight but lack statistical significance. The test needs enough volume in both groups and enough time to account for natural conversion variation.
The second mistake is testing once and treating the result as permanent. Incrementality changes based on market conditions, audience saturation, creative fatigue, and competitive activity. A channel that tested at 30% incremental lift 6 months ago may test differently today. Regular re-testing builds a more reliable picture than a single data point.