What Is Incrementality Testing? Measuring What Ads Actually Cause
Incrementality testing is an experiment that measures how many conversions were actually caused by your advertising, by comparing a group exposed to the ads against a comparable group that was not. The difference is the incremental lift.
Attribution reports assign credit to ads that were seen or clicked, but some of those customers would have bought anyway, for example loyal buyers targeted by retargeting. Incrementality asks a different question: what would have happened without the ads? To answer it, you hold out a control group (see what-is-a-control-group-in-ad-testing) through audience splits, geographic holdouts or platform lift-test tools, then compare conversion rates.
If the exposed group converts at 5% and the holdout at 4%, the lift is one percentage point and only a fraction of attributed conversions were truly incremental. This is most useful for channels where attribution tends to over-credit, such as branded search and retargeting.
It has costs. You need enough volume for a result you can trust (see what-is-statistical-significance-in-ab-testing), the holdout means forgoing some exposure, and results describe a particular period and setup rather than a permanent truth. Small businesses often lack the volume for clean geographic tests, so it is typically a tool for larger spend, with simpler before-and-after checks as a rougher substitute.
Why it matters
Infinall AI produces research, strategy, ad copy and creative that land in an Approval Queue for review before export. It does not run lift studies, split audiences, measure conversions or buy media, so incrementality testing happens in your ad platforms or measurement tools. Where it can help is generating the distinct creative variants you might want to test against each other, which you approve before anything is exported. Be wary of any tool that claims to prove incrementality without a real holdout design.
Related terms
Frequently asked questions
How is incrementality different from attribution?
Attribution distributes credit among touchpoints a customer encountered. Incrementality tests whether those conversions would have happened anyway by comparing against a group that did not see the ads. Attribution describes what happened along the path; incrementality estimates what the ads caused.
Can a small business run an incrementality test?
It is harder, because reliable results need enough conversions in both groups. Some ad platforms offer built-in lift tools with minimum spend requirements. With low volume, simpler approaches such as comparing sales in matched periods or regions can give a rough signal, but treat the conclusion cautiously.