Measurement and evidence
What is incrementality in advertising?
Free, with no card. We read your homepage, draft who buys from you and show your monthly price before you join the waitlist. Nothing is bought until you've approved the plan and the ads.
what it means here
Here is what the term means for a UK buyer
The definition above holds anywhere. Buying media in the UK adds the detail below, including where AdBuyMCP does something different.
Sold by most platforms in this category and refused by almost none, which is the difference worth caring about. AdBuyMCP runs a matched-market geo-lift test and puts a power analysis in front of it at 5% significance and 80% power. If the design cannot detect a plausible lift, it returns an underpowered verdict, the required sample days and the minimum detectable effect, and reports no lift figure, because a number from an underpowered test would be noise. Nothing here has produced a causal result for a customer, because nothing has run live.
How do you measure incrementality?
Compare an exposed group with a comparable group that was deliberately held back, and read the difference. In media the practical design is a geo-lift test: buy some markets, hold out others that behaved alike beforehand, and compare how each changed. It needs no login, pixel or match rate, so it works on television, cinema and out-of-home, where person-level tracking is impossible.
The alternatives, marketing mix modelling and attribution, produce estimates and correlations. Both are useful, and both should carry those labels when they reach a board pack.
Why do most incrementality tests fail?
Two failures account for most of them, and both are checkable before the money goes out. The first is power: a design too small to detect a plausible lift returns a random number that looks like a result. The second is contamination: control markets that received some of the campaign make the lift look smaller by an amount nobody can know.
AdBuyMCP checks both before a readout. An underpowered design returns the sample days it would need, and a contaminated one returns the names of the markets that received delivery, so you learn what to change while the plan can still change.
Type your company website or work email to see who we think buys from you and what reaching them 3, 5 or 12 times a week would cost.
Free, with no card.
related terms
These terms sit next to this one
Most confusion in media buying comes from two adjacent terms being used interchangeably, so these are the neighbours worth reading next.
Geo-lift test
An experiment that runs advertising in some geographic markets and withholds it from comparable others, then compares outcomes between them. It can measure what the advertising caused without a login, a pixel or a match rate.
Holdout
A group deliberately excluded from seeing a campaign and kept as a comparison against the group that was exposed to it. Without one, the counterfactual (what would have happened anyway) has to be assumed rather than observed.
Statistical power
The probability that a test will detect a real effect of a given size, if an effect of that size genuinely exists in the population being measured. A design with low power will usually miss a genuine effect, raises the share of the results it does produce that are noise rather than signal, and exaggerates the size of the effects it does detect.
Attribution
Assigning credit for an outcome to the advertising touchpoints that preceded it, according to a rule the advertiser chose. It describes correlation in a conversion path, and it cannot show on its own that the advertising caused anything.
same group
These terms are also in Measurement and evidence
What you can prove afterwards, and how to tell a number that means something from one that does not.