How incrementality works
Incrementality testing answers the question: “Did this ad campaign generate revenue that wouldn’t have happened without it?”
The problem with attribution
Standard attribution (last-click, first-click) tells you which ad a customer clicked before converting. But it doesn’t tell you whether they would have converted anyway. A customer who Googles your brand name and clicks a brand ad would likely have bought without the ad — attributing that sale to the ad overstates its value.
Incrementality testing measures the true lift — the additional revenue caused by the ad, not just correlated with it.
How Active Reach measures it
Active Reach uses geo-holdout experiments — a proven methodology used by major platforms:
- Split regions — divide your target geography into test (ads run) and control (ads paused) groups
- Run the experiment — keep both groups running for a measurement period
- Compare outcomes — measure conversions in both groups
- Calculate lift — the difference between test and control is your incremental impact
The test group sees your ads. The control group doesn’t. Everything else stays the same — organic marketing, pricing, product availability.
Setting up an experiment
Go to Ads → Proof → New experiment:
Select the campaign to test
Pick an active ad campaign you want to measure.
Define test and control regions
Active Reach suggests a geographic split based on your audience distribution. You can adjust:
- Test regions — where ads continue running
- Control regions — where ads are paused for the experiment duration
- Split ratio — typically 80/20 or 70/30 (test/control)
Set the measurement period
Recommended: 2-4 weeks. Shorter periods have more noise; longer periods are more reliable but delay decision-making.
Launch
Hit Start experiment. Ads in control regions are automatically paused. The experiment runs until the measurement period ends.
Reading results
After the experiment completes, the results page shows:
| Metric | Description |
|---|---|
| Incremental conversions | Additional conversions caused by the ad (test minus control, normalized) |
| Incremental revenue | Additional revenue from those conversions |
| Lift percentage | How much the ad increased conversions vs. no-ad baseline |
| Statistical significance | Confidence level that the result isn’t due to random variation |
| Cost per incremental conversion | Your true cost to acquire one additional customer |
A statistically significant positive lift means the campaign is genuinely driving new business. A non-significant or negative result means the ad budget might be better allocated elsewhere.
What’s next
- Budgets & governance — reallocate budget based on incrementality results
- Analytics overview — workspace-level performance metrics