Skip to content

Data and measurement

Your guide for measuring true media impact with incrementality

As consumers engage with brands across more channels and devices, marketers have more opportunities to connect with them throughout their path to purchase. But those journeys are also harder to follow, and last-touch attribution, which credits only the final interaction before a conversion, may not show which touchpoints actually drove the result. Incrementality, also known as lift testing, helps fill that gap by answering a more important question: Did advertising actually cause the outcome?

Illustration of a compass surrounded by bar graphs, pie charts, and other data visualizations.

Share:

Performance marketers, such as travel and insurance marketers, often prioritize quick conversions through lower-funnel tactics like retargeting and search ads, traditionally measured with last-touch attribution. For example, if a traveler clicks a paid search result, then immediately books a flight, last-touch attribution would claim their search was responsible for the booking. But this approach is ripe to change, as consumer journeys become more fragmented and harder to track, and as businesses prioritize real growth over media performance. Because after all, receiving credit for an outcome isn’t the same as causing it. As you seek to improve measurement accuracy and plan for longer-term growth, last-touch limitations become harder to ignore.

Beyond last-touch attribution

Incrementality helps answer how much impact your media has on outcomes by comparing what happens when people are exposed to an ad and when they aren’t. This gives you a clearer way to understand the true business impact of your campaigns.

Consumer journeys have become harder to follow as people move between devices and media channels, interacting with multiple brands and platforms as they research a purchase. Combined with the reduction in verified first-party user data on the open web, deterministic measurement, which ties ad exposure to outcomes through verified identity matches such as email addresses, has become more difficult. 

Last-touch attribution also under-credits channels that are not ‘clickable’, such as CTV, even though it may drive significant additional demand. Meanwhile, a channel like paid search may appear to be a top performer because it receives credit for conversions that would have happened regardless of ad exposure.

Given these changes and limitations, adding lift testing to your strategy can be valuable. Lift testing helps measure true business impact, and because it uses a modeled methodology, it is more immune to changes in user online behavior. 

Consider two people who see the same ad and ultimately make a purchase. One was already planning to buy and simply used a sponsored link as shortcut. The other would not have purchased without seeing the ad first. An attribution model may give the campaign credit for both purchases. 

Incrementality, however, asks: “How many more purchases occurred because the campaign ran?” That’s important because the conversions a campaign receives credit for are not always the conversions it created. When measurement follows the final touchpoint too closely, the lower-funnel media channels can look more valuable than the ones that helped create demand in the first place.

Measuring incremental growth 

Incrementality considers what would have happened if an ad had not been shown. By creating a comparison between people who see advertising and those who don’t, incrementality gives you a way to look beyond which touchpoint received credit and understand whether your media created additional outcomes.

Case study

In one campaign, Marriott employed a mix of CTV, digital out-of-home, online video, digital audio, and display to reach travelers across different stages of the booking journey. People exposed across more than one media channel were 1.8 times more likely to visit Marriott’s website than those reached through a single channel. If a traveler later searched for Marriott and booked, however, a last-touch model would only give search the credit because it was the final recorded interaction, making the influence of earlier touchpoints harder to recognize. 

For a travel brand, that could mean testing for incremental bookings, revenue, new customer acquisition, or loyalty sign-ups. Other brands may have different outcomes, but the principle is the same: When you start measuring the outcomes that happen because of media exposure, rather than simply after media exposure, you’ll be able to make more informed planning decisions on where to increase investment, reduce spend, and come up with new strategies to test.

Choosing the right measurement method  

Incrementality is a powerful way to learn what strategies move the needle for your brand, but it shouldn’t replace every other form of measurement. Often a combined approach is best, with each one doing the job it is best suited for. 

  • Last-touch attribution: Identifies the final touchpoint before conversion, providing directional insights on what drives conversions in short sales cycles. Results update in near real time and, with Koa Optimizations enabled, can inform your in-flight optimizations.

  • Incrementality: Uses randomized experiments to help control for other factors and isolate media’s influence on generating business outcomes. This can give you a clearer understanding of the value that each campaign, channel, or targeting strategy brings, helping inform your future planning.

Incrementality is especially valuable because its results can help validate and inform other measurement methods. If a lift test shows that a campaign generated substantially more or less incremental impact than an attribution model suggested, you can use that evidence to improve how you interpret attribution and make planning decisions. Or for brands that run Marketing mix modeling (MMM) to evaluate all marketing activity and allocate future budgets, lift testing provides a ‘ground truth’ to anchor analysis and creates a common metric that can be used to compare all channels and activations.

Using multiple methods together can provide a more complete understanding of how your media is performing, from the signals that can guide real-time optimization to the broader impact of your investments over time.

Measurement informs decisions

A lift study is most useful when it answers questions your brand needs to solve. For example:

  • Should we increase investment in this media channel? 

  • Are we reaching people who would not have converted otherwise? 

  • Is a new audience generating incremental sales? 

  • Are we approaching diminishing returns? 

  • Would expanding into another media channel create additional growth? 

Incrementality in action

See how Tripadvisor and PMG tested three audience targets against a control group and found a 2.7x lift in conversion rate and 75% better cost per booking.

Once you receive a result, you can use it to inform your next study, turning incrementality into an ongoing system for learning and decision-making, whether that’s testing a new audience, like Tripadvisor did, reallocating investment, expanding into another media channel, or asking a new question.
 
The best way to measure lift depends on your business question as well as the data available. The Trade Desk supports multiple approaches to measuring incremental impact, including our own conversion lift and geo lift solutions, or third-party partner offerings. 

Together, these capabilities can help make incrementality part of an ongoing measurement strategy, giving you a clearer understanding of campaign impact and helping inform what to do next to continue growing. 

Want to learn more about how incrementality can help you understand the true impact of your media investment? Reach out to your Trade Desk representative to explore the measurement approach that makes sense for your business. 


This information is provided solely for background and is not a representation or guarantee of any future performance.