Glossary · MeasurementTOFU
Incrementality
The short answer
Incrementality is the portion of results that happened because of your marketing and would not have happened without it. A retargeting campaign may be credited with 500 sales, but if 400 of those customers would have bought anyway, its incremental contribution is 100.
It's the question attribution can't answer: not "who was present?" but "what did we cause?"
How incrementality is measured
- Geo holdout — pause or reduce a channel in some regions, keep it running in comparable ones, and compare total sales.
- Conversion lift studies — platform-run experiments that withhold ads from a random control group.
- Scheduled pauses — turn a channel off for a defined period and watch total revenue, carefully controlling for seasonality.
- Marketing mix modelling — statistical estimates across many periods and channels.
What incrementality tests usually reveal
- Retargeting is often less incremental than platform reports suggest
- Brand search protects demand but creates little on its own
- Prospecting and video are often more incremental than last-click shows
- Coupon affiliates frequently intercept sales rather than create them
Incrementality calculation example
| Group | Customers | Conversions | Rate |
|---|---|---|---|
| Exposed to ads | 100,000 | 2,400 | 2.4% |
| Holdout (no ads) | 100,000 | 1,800 | 1.8% |
Incremental lift = (2.4% − 1.8%) ÷ 1.8% ≈ 33%. Of the 2,400 conversions, about 600 were incremental. See attribution modelling.
Incrementality in practice
A brand pauses Meta prospecting in a few comparable regions for four weeks and compares sales with regions where it continued. The difference shows how many sales the ads actually caused.
Ways to measure incrementality
- Geo holdout tests
- Platform conversion lift studies
- Time-based tests
- Media mix modelling
Common incrementality mistakes
- Tests too short or too small
- Ignoring seasonality between test and control
- Treating one test as permanent truth
Frequently asked questions
How often should we run incrementality tests?
For major channels at meaningful spend, once or twice a year and before any large budget shift.
Do small brands need incrementality testing?
Below a certain spend, the noise overwhelms the signal. Use blended MER and simple before/after comparisons instead.
What is a geo holdout?
A test where a channel runs in some regions and is paused in comparable regions, then sales are compared.
Can platforms measure incrementality?
Meta and Google offer conversion lift studies for eligible advertisers.
Why does incrementality matter?
Because attribution shows correlation, not causation — incrementality shows real impact.
How often should incrementality be tested?
Periodically — especially before major budget changes.
Related terms
Full glossaryTalk to a strategist
Thirty minutes. Your numbers. A straight answer.
Book a strategy call with the people who'd actually run your account. We'll tell you what we'd do — or that you don't need us yet.