Incrementality Testing: What’s Beyond First-Touch and Last-Touch Attribution

Ever Feel Lost Measuring Marketing Impact?

Have you ever felt lost trying to figure out the real impact of your marketing without relying on tracking a user's activity—like first-touch or last-touch attribution? You are definitely not alone. Many marketers face this exact challenge. We all know that spending on brand marketing or other hard-to-measure initiatives can boost awareness, drive engagement, and increase sales. But if we can't show a clear return on investment (ROI), it becomes difficult to secure the budget to continue these efforts.

The reality is, all marketing should be able to demonstrate that it helps drive sales. If a brand campaign isn't generating sales, we need to reevaluate it and consider making adjustments. The challenge, though, is that some of the most impactful marketing activities are also the hardest to measure using traditional methods. Just because something is hard to track doesn’t mean it lacks value, and it certainly doesn't mean we should avoid investing in it.

So, how can we effectively measure the impact of these harder-to-track channels? Let me introduce you to a powerful approach: incrementality testing.

What is Incrementality Testing?

Incrementality testing involves showing your marketing to one group of people and not showing it to another group—while keeping everything else the same. For example, you might run ads to one group of potential customers while withholding those ads from another similar group to see if there is a difference in conversions.

This is the key: everything else needs to stay the same so that the only difference is whether or not they saw your marketing.

Take, for instance, Group A, which sees your YouTube ads, while Group B (the control group) does not. If Group A has a higher conversion rate compared to Group B, and all other factors remain the same, you can confidently conclude that those ads are making a difference. Incrementality testing helps you isolate the impact of a specific marketing channel. This is crucial because it helps you understand which channels are truly driving results, allowing you to allocate your budget more effectively.

The Power of Controlled Experiments

Incrementality testing is the most scientific way for marketers to prove a cause-and-effect relationship between a marketing activity and a business outcome—not just a correlation. It’s similar to a clinical trial: a controlled experiment in which only certain variables are changed while everything else remains constant. This helps ensure that the results are accurate and unbiased.

Think of it like this: imagine you see an ad on YouTube but don’t click on it. Later, you search for that company on your computer and sign up for a free trial. With traditional attribution models, there wouldn’t be a way to connect that YouTube ad to your later sign-up. Your finance team and CEO might ask, "What’s the ROI on YouTube ads?" and you wouldn’t have a solid answer. Incrementality testing solves this problem by comparing the audience that saw YouTube ads to a group that didn’t, allowing you to prove the ads' value.

How Primer Can Help

Primer gives you the tools to set up these types of experiments. You can create a control group—some people in your audience who won’t see your ads—and compare their results to those who do see the ads. We work to make sure these groups are as similar as possible to ensure that the results are both accurate and unbiased.

Example incrementality report

Imagine running YouTube ads, and the group that saw the ads has 50 conversions while the control group has just 10. By using statistical analysis, such as a chi-square test, you can demonstrate that this difference is meaningful—proving that the YouTube ads are making a real impact, even if people aren't clicking on them directly.

A Scientific Approach in a Changing World

Controlled experiments like incrementality testing help you measure the impact of marketing channels that are harder to track—something that becomes increasingly important as privacy regulations tighten. With more privacy rules in place, traditional tracking methods are becoming less reliable. But that doesn’t mean we have to fall back on saying, "Trust me, it’s working." Instead, we can use scientific methods to accurately determine the true impact of our campaigns.

If you want to see the real impact of channels you’ve been hesitant to explore or find hard to measure, give Primer a try. Reach out to our team—we’d love to help you prove the value of your marketing investments.