Incrementality calculator.
Did your ads actually cause those sales? Compare people who saw your ads with a control group who didn't, and see the extra conversions your advertising really drove.
1. Group that saw your ads
2. Control group that didn't see your ads
3. Optional: to see what the extra sales cost
Lift = (test conversion rate − control conversion rate) ÷ control conversion rate. Extra conversions = the difference in conversion rate × people who saw ads. Check the result is reliable with our A/B test significance calculator.
Free, no sign-up. Everything is calculated in your browser.
How to measure incrementality.
Ad platforms report every conversion that happens after someone clicks or sees an ad, including people who would have bought anyway. Incrementality strips those out and asks a simpler question: how many extra sales did the ads cause?
To find out, you compare two similar groups: one that sees your ads (the test group) and one that doesn't (the control or holdout group). The difference in their conversion rates is the effect of your advertising.
The formulas
- Conversion rate = conversions ÷ people, for each group.
- Lift = (test rate − control rate) ÷ control rate.
- Extra (incremental) conversions = (test rate − control rate) × people in the test group.
- Cost per extra conversion = ad spend ÷ extra conversions.
- Incremental ROAS = (extra conversions × average order value) ÷ ad spend.
A worked example
100,000 people saw your ads and 1,200 bought (1.2%). A control group of 20,000 people saw no ads and 200 bought (1.0%). That's a 20% lift, and 0.2% × 100,000 = 200 extra sales caused by the ads. With £10,000 of spend, each extra sale cost £50, and at a £60 average order, incremental ROAS is 1.2x. Ads Manager might have credited all 1,200 sales; only 200 were truly incremental.
How to run a test
- Platform lift studies: Google, Meta, TikTok, LinkedIn and Microsoft can create the test and control groups for you. Our lift test guide covers each one.
- Geo tests: switch ads off in comparable regions and compare sales against the rest.
- Everyday estimates: settings such as Meta's incremental attribution model incrementality in normal campaigns.
Whatever the method, judge the result on business outcomes: the principle behind commercial alignment over channel metrics. Want a test designed and read out for you? See our incrementality testing service.
Keep calculating.
A/B test significance calculator
Check whether the difference between two variants is statistically significant before you call a winner.
Open tool →Testing & measurementA/B test sample size calculator
Find out how many visitors you need, and how long to run your test, before you start, so you don't call a winner too early.
Open tool →TrackingTracking checker
Enter your website and see which pixels and tags it runs, what fires before cookie consent, whether Consent Mode v2 and ad click IDs work, and if you have server-side tracking.
Open tool →Incrementality questions.
01What is incrementality in advertising?
Incrementality is the extra result your ads cause, the sales or leads that would not have happened without them. It is measured by comparing a group that saw your ads with a similar group that did not.
02What is a holdout test?
A holdout (or control) group is a set of people or regions deliberately not shown your ads. Comparing their conversion rate with the group that did see ads shows how much your advertising really adds.
03What is incremental ROAS?
Incremental ROAS is the revenue from incremental conversions divided by ad spend. It is usually lower than platform-reported ROAS, because it excludes sales that would have happened anyway, but it is a truer measure of what your ads earn.
04How big should my control group be?
Big enough to give a reliable conversion rate, usually at least 10–20% of your audience, and enough people to record a meaningful number of conversions. Small control groups produce noisy results.
05Can I run an incrementality test on Meta or Google?
Yes. Meta and Google both offer conversion lift studies, and you can also run geo tests by switching ads off in comparable regions. Meta’s incremental attribution setting uses models to estimate incrementality in everyday campaigns.
We measure paid media on incremental revenue, not platform credit, from test design to reading the results.