Incrementality: The Best Way to Measure Ad Success

Measuring Ad Performance With Incrementality

Return On Ad Spend, or ROAS, relies on a very simple formula for calculating the financial benefit of advertising campaigns: Revenue / Cost. Compared to the science experiment-esque process that is calculating incrementality, it’s simpler, more digestible, and one of the best ways to measure ad performance.

Incrementality provides valuable information where ROAS falls short. Calculating profit by dividing all ad revenue by all advertising costs, though a simpler calculation, misses much of the point—it lumps together all advertising campaigns and all users without recognizing the complexities of individual ad campaigns on individual user groups. With incremental lift, advertising campaigns can be broken down and examined individually for their effectiveness, allowing advertisers to better target spending.

So, which is the most effective overall? Before we answer that question, let’s dive into what incrementality is and how it’s calculated.

What exactly is incrementality?

Basically, incrementality refers to the measure of the incremental lift—or the likelihood of a consumer to convert—that advertising spend provides to the overall conversion rate. In other words, incrementality provides the percentage of conversions that were received as a direct result of an advertising campaign.

Think of it like this: when a new ad campaign for a direct to consumer monthly subscription box—let’s reference it here as Company X—is released to the public, calculating incremental lift can define what percentage of people see that ad and decide to sign up for a subscription. ROAS, on the other hand, only calculates the total amount of revenue that offsets the cost of advertising campaigns but not an ad campaign’s success in conversions. incremental lift looks more specifically at this information to determine the effectiveness of individual advertising campaigns—how well they worked, what sales revenue would have been without the ad campaign, etc. According to the Nanigans blog: “[u]nderstanding incrementality allows advertisers to optimize their ad budgets by allocating retargeting ad spend in the most impactful way possible.”

Incrementality is calculated by comparing two groups: those who were exposed to an advertising campaign vs. a control group of users who were not exposed to it. After observing the conversion rates of both groups, incremental lift can be calculated through two formulas:

Lift = Test (conversion rate) – Control (conversion rate) / Control (conversion rate)

Incrementality = Test (conversion rate) – Control (conversion rate) / Test (conversion rate)

Let’s say Company X decides to observe incrementality for a new retargeted ad campaign. After comparing the purchase rate increase between the test group and the purchase group, they determine an incremental lift rate of 45% for the test group after exposure to the retargeted ad. From this information, we can see that they received 45% more conversions to their shopping service as a direct result of running the retargeted ad campaign. Because of running this specific retargeted ad campaign, Company X was able to increase the amount of conversions to their product by 45%.  Had they not run this retargeted ad campaign, the conversion rate would be decreased by 45%.

The Importance of Incrementality

This information is valuable for Company X, as it can show which ad campaigns are more or less successful than others. However, calculating incremental lift can also give insight into where retargeted spending should be prioritized. For instance, say Company X, after observing and calculating incrementality for the same retargeted ad campaign, discovers an incremental lift rate of 15% rather than 45%. This could be a result of a failed ad campaign, but it could also suggest that the group of users studied were likely to have converted whether they were exposed to an ad campaign or not. Therefore, spending on retargeted ads should not be prioritized for this group, as they were not conclusively affected by the ad campaign in the first place. Instead, this spending should be targeted towards groups with much higher incrementality percentages. By correctly interpreting information like this, Company X can better plan its advertisement spending to reach the most receptive audience possible.

For obvious reasons, incrementality has become a very popular method for measuring the effectiveness of ad campaigns. However, for it to be most advantageous, it is crucial to consider errors like premature decision-making, or misinterpretation of data. For instance, factors such as seasonality can skew data drastically. If, for example, a company calculates the incremental lift of an ad campaign run on Cyber Monday, the control group response rate would likely be much higher than it would be normally; this would lead to a lower incremental lift percentage than might otherwise have appeared. This is because on Cyber Monday, consumers are more likely to complete purchases anyway, regardless of their exposure to an advertisement. We’re all so bombarded by the idea of the deals on Cyber Monday, we don’t care whether we see an advertisement for JCrew blazers or appliances at Target, we’re going to (electronically) throw down a credit card anyway. As a result, when calculating incrementality around this holiday, the lift for this data would be misleadingly small.

Not taking this factor into account and subsequently making crucial decisions to change advertising strategy would be a critical mistake–an otherwise successful ad campaign could be discarded, advertising spending could be misallocated, or an entire group of users could be incorrectly labeled as less likely to respond to advertisement campaigns. However, by taking the impact of Cyber Monday into consideration, and subsequently recalculating incrementality on a more neutral buying day, companies can get a better, truer understanding of an advertising campaign’s impact on consumers.

To conclude, for those who want only a summary of information regarding the impact of their ad campaigns, ROAS is a sound tool. However, for those looking for more—incrementality offers a much closer look, helping advertisers better understand the individual impact of ad campaigns and make better informed decisions based on this knowledge. More and more, companies are drawn toward measuring their ad performance with incrementality, and for good reason. Moral of the story: don’t be afraid to put in a little more work to get better, more in-depth information; it will pay off in the long run.