Stockertim Dictionary · Marketing
Incrementality
Also called incremental lift, uplift, incremental impact
In short
The extra result a piece of marketing caused, found by comparing what happened with it against what would have happened without it.
- Intermediate
- Concept
- Checked
In detail
Some of the people who see an advert and then buy would have bought anyway: they were loyal customers, they were already searching or the season was in their favour. Incrementality is the part of the result that exists only because of the marketing. It answers the question that attribution and platform reports cannot: what did this activity add?
What would have happened without the activity is called the counterfactual, and it can never be observed directly for the same people at the same time. It is estimated instead from a comparable group that did not receive the activity, a control group, or from comparable places or periods. The difference between exposed and control results is the uplift.
Uplift can be expressed in absolute terms, as extra sales or a gap in percentage points between conversion rates, or in relative terms, as a share of the control result. Dividing the cost of the activity by the incremental results gives the cost per incremental sale, usually far higher than the cost per sale a platform reports, because the platform counts every sale it touched.
Reliable measurement needs groups that are alike before the test, enough people or places to detect a realistic uplift, no stray exposure of the control group and no stopping early when results look good. Results are reported with their uncertainty, since a small uplift measured on modest groups may be real but imprecise.
Formula
Incremental results = (test rate − control rate) × test group size
- test rate
- the conversion rate of the group that saw the activity
- control rate
- the conversion rate of the matched group that did not
- test group size
- the number of people in the exposed group
Worked example
Zuma Water shows a video campaign costing £6,000 to a random half of 80,000 customers. The exposed half converts at 3.0% and the control half at 2.5%.
- Exposed sales: 40,000 × 3.0% = 1,200
- Control sales: 40,000 × 2.5% = 1,000
- Incremental sales: (3.0% − 2.5%) × 40,000 = 200
- Cost per incremental sale: 6,000 / 200 = £30
The campaign added about 200 sales at £30 each, even though 1,200 people who saw it bought.
In practice
Advertisers run holdout tests on email and app messages, matched-region tests for television and outdoor, and platform-run lift studies, then use the incremental cost to decide how budgets are split between channels.
Watch out
- A reported cost per sale is not the same as the cost per incremental sale.
- Before-and-after comparisons confuse the campaign with seasons, prices and rivals' moves.
Learn it properly
- Experiments and incrementality5 min lesson Free course: Marketing analytics and measurement
- Inflation-linked bonds in depth: indexation, breakevens and their hidden risks15 min lesson Free course: Bonds & fixed income: Advanced
- Testing advertising and judging effectiveness5 min lesson Free course: Advertising
Official sources
- The Magenta Book: guidance on evaluation · HM Treasury
- What is experimental design? · National Institute of Standards and Technology
- Evaluate and learn: the Evaluation Cycle · Government Communication Service
Facts last checked on 3 November 2026.
Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.