Incrementality is the extra result, such as sales or sign-ups, that your advertising causes beyond what would have happened without it. In Meta ads, it separates people who bought because of an ad from people who were going to buy anyway.
Credited versus caused
Standard reporting credits an ad with every conversion that follows a click or view inside the window. An attribution model decides who gets that credit. It can't tell whether the person would have bought without the ad. Incrementality answers that, by comparing people who could see your ads with a similar group who couldn't.
Two ads can show the same credited results and very different incrementality. Retargeting people who were already on their way to checkout often claims many conversions that would have happened anyway. Reaching new people who never heard of you may show fewer credited conversions and a larger real effect.
Let Madgicx's AI do the work
Its AI marketer finds what's working and prepares the next move. You approve.
Try for Free ($0 Trial)How to calculate incrementality
Incremental conversions = test group conversions − control group conversions
Lift = incremental conversions ÷ control group conversions
Take two equal-sized groups. The test group can see your ads and the control group can't. If the control group converts at 1.0% and the test group at 1.2%, the extra 0.2 points are the ads' effect. Divide your spend by incremental conversions to get a cost per incremental conversion, which can be a very different number from the cost per result in Ads Manager.
How Meta measures it
Meta's Experiments page describes two kinds of lift test:
- Conversion lift "enables you to measure the incremental performance of your Meta ads."
- Brand lift measures "the incremental effect your advertising has on brand awareness, perception, recall or purchase intent." See brand lift for how it works.
In both, two audiences are created: a test group that has the opportunity to view your ads and a control group that does not see your ads. Meta also says there is no additional cost to create a test.
Meta's separate incremental attribution setting works differently. It optimizes delivery for conversions a model predicts the ad caused. A lift test measures the effect directly against a group that didn't see the ads. See the attribution window entry for Meta's attribution settings.
How to run a test that you can trust
- Pick one question. For example: do prospecting ads increase purchases?
- Hold out a control group that doesn't see the ads. Informal tests, such as switching campaigns on and off, mix in seasonality and are hard to read.
- Keep everything else steady. Changing creative, budgets or offers during the test muddies the result.
- Run it long enough for the conversions you care about to happen, and for the result to settle.
- Compare cost per incremental conversion with your return on ad spend targets, and act on the gap.
Other ways to measure incrementality include geo experiments, where some regions get ads and others don't, and marketing mix modeling, which works from aggregate spend and sales. Multi-touch attribution shares credit across a path, and does not by itself measure incrementality.
Example. A shoe brand tests its Meta prospecting campaign with a conversion lift study. The test group and the control group each hold 500,000 people. The test group makes 6,000 purchases (1.2%) and the control group 5,000 (1.0%). That is 1,000 incremental purchases and a lift of 1,000 ÷ 5,000 = 20%. The brand spent $30,000, and Ads Manager credited the campaign with 4,200 purchases, or $7.14 each. Per incremental purchase, the cost is $30,000 ÷ 1,000 = $30. With an average order value of $110, $30 per extra sale still pays, so the brand keeps the campaign. It does stop using the $7.14 figure to forecast growth.
In Madgicx. Madgicx Cloud Tracking sends conversions to Meta from the server, so Meta sees and attributes more of them, and dashboards set Meta's credited results next to store revenue. The comparison with revenue is a useful check next to a lift test.
FAQ
What is incrementality in marketing?
It is the portion of your results that your marketing caused, over what would have happened without it. It is measured by comparing a group exposed to ads with a similar group that isn't.
What is an incrementality test?
It is an experiment with a test group that can see your ads and a control group that can't. The difference in conversions between them is the incremental effect.
What is the difference between incrementality and attribution?
Attribution shares credit for conversions among the touchpoints before them. Incrementality measures whether the ads caused those conversions at all. The two answer different questions and work best together.
Does Meta offer incrementality testing?
Yes. Meta's Experiments tool includes conversion lift and brand lift tests, with test and control groups. Meta says there is no additional cost to create a test.
What is incremental ROAS?
It is the revenue the ads caused divided by what you spent on them, rather than the revenue the platform credited. It is usually lower than credited ROAS, because it leaves out sales that would have happened anyway.
Related
- Related terms: Brand lift, Attribution model, A/B testing
- In depth: Facebook ads lift measurement
Sources
- About Experiments, Meta Business Help Center
