Marketers have always wanted to know whether the time, money, and effort they had put into marketing activities are causing users to convert. To find the answer, you need to be able to distinguish the extent of organic traffic from the impact of your paid ad campaigns.
The catch is that traditional attribution alone is not sufficient for causal inference, thus telling a fragmented story, if not inaccurate. Moreover, there can be additional issues related to budget and marketing mix resulting from incomplete marketing performance measurement.
Then how do you prevent such chaos and take a step closer to success? Incrementality can show you the way.
Incrementality is the measure of supplemental business outcomes driven by specific marketing efforts. By identifying the causal relationship between touchpoints and conversions, marketers are able to determine which events would not have occurred in the absence of a particular interaction. In other words, incrementality demonstrates the true power of each component of your marketing mix that is otherwise difficult to isolate and assess.
While the concept of incrementality is rooted in science, in the marketing world, it is essentially used to give an understanding of ad effectiveness. This allows marketers to deal with the two essential issues: justifying ad spend and increasing budget efficiency.
1) Validating your marketing budget
Profit is inarguably the ultimate business objective. To get there, marketers need to make sure that the return on investment is high and that such a result was driven by marketing activities. The problem is that the Last-Touch Attribution (LTA) model, which is the industry standard today, has limitations in capturing causal effects because it allocates 100% of the credit for a conversion to the single last touchpoint. The model oversimplifies the complex user journey and ignores the contribution of many earlier touchpoints as well as users’ latent inclination to convert.
Thankfully, incrementality measurement could be a way to solve the limitations of LTA, highlighting the significance of marketing campaigns in conversion generation. If you can prove that the conversion would not have happened without your marketing efforts, you can explain why your company needs to invest in marketing.
2) Optimizing your marketing mix
Incrementality also helps with channel-level analysis. By identifying the higher-performing channels, you can focus on ad platforms that bring tangible results, thus preventing budget waste and improving marketing efficiency.
There is always an overlap between organic traffic and paid conversions, but the separating line is often blurry, leading to less-informed decisions. This is where incrementality measurement comes in, to reveal which of your marketing components are creating actual value. In specifics, it provides answers to the following questions:
Working with incrementality allows you to get the most accurate, actionable insights into the impact of your marketing activities. Not only can you see the causality between touchpoints and conversions, you can clearly determine the best set of paid ads across various channels.
There are several ways to measure incrementality, but experimental and observational are the principal approaches. An experimental study divides users into a test group and a control group to compare the performance of the two groups and measure the lift, whereas an observational study uses historical data to draw causal inferences.
The experimental method produces more accurate results, but it involves difficulties as well. For instance, there must be data science experts on your team to create two groups that share similar characteristics. Still and all, no matter how carefully planned, it is almost impossible to make the groups perfectly identical. To handle such a situation, you can get help from Meta and Google’s Randomized Controlled Trials (RCTs) for incrementality testing, which randomly allocate users to two groups.
Yet, experiments inevitably entail another drawback – time, cost, and resources. Turning off ads to the control group during the study implies a lost opportunity for sales. Moreover, it takes time to confirm and analyze the results, meaning that applying yesterday’s data in today’s decision making is not an option. This is why many martech tools like Airbridge have chosen to work with the observational method, providing machine learning-driven insights into user behavior.
Top ad platforms Meta and Google offer lift measurement tools that control the delivery of ads to groups with similar attributes. Based on real-world experiments and data, not only are the results standardized, accurate, and reliable, but they also effectively show the causal effect. Let’s dig deeper into how each ad platform evaluates incrementality.
Meta’s lift tests help you see the true value of your Facebook ads. This experiment splits your chosen audience into groups who did and did not see your Facebook ads to understand their causal impact on specific business objectives such as brand awareness or sales. The control group here would be those who match your audience but are intentionally kept away from your ads.
There are three kinds of lift tests you can run on Meta:
👉 Learn more about Meta's lift tests
Google’s lift tests measure the effectiveness of your Google ads, which you can use to adjust and improve your campaigns. Since it is not available for all Google Ads accounts, you need to contact your Google account representative to measure incrementality on Google.
There are two kinds of lift tests you can run on Google:
👉 Learn more about Google's lift tests
As explained earlier, the observational method can be an alternative to the experimental method which requires time, cost, and resources. However, when using historical data, it is impossible to randomly allocate the users into treatment and control groups.
To overcome such a challenge, matching is widely used in the industry. By finding similar observable characteristics between two user groups and excluding all unmatched users from the causal analysis, matching enables a comparison of outcomes while minimizing selection bias caused by covariates.
Propensity Score Matching (PSM) is another popular technique that uses scores to estimate the probability of a user being exposed to a particular ad. After matching users with similar propensity score values from the test and control groups, all unmatched users are discarded and the actual impact of an ad is found by comparing the performance of the two groups.
It is every marketer’s goal to get the most truthful and complete insights into marketing performance and drive growth. While LTA has been highly in demand thanks to its convenience, it is neither a perfect nor the only way to measure marketing performance, especially when it comes to evaluating the true power of your marketing activities.
Incrementality measurement gives a holistic perspective for identifying the most effective campaigns and optimizing paid ad strategies. If you would like to find out the real value of your marketing, contact our team of experts to discuss. You can also check out our white paper to take a deep dive into the fundamentals of incrementality measurement.