It’s table stakes to measure the outcome of an ad campaign. How did it shift awareness and perception of a brand? What actions, like a click or purchase, did it drive? 

But how do we isolate a campaign’s outcomes, with confidence?

No attribution signal or metric is ever 100% deterministic. That is, while we can strongly associate a shift in perception or an action with exposure to an ad, it can be more difficult to judge the full extent of influence the ad had on the outcome. That’s because humans are complex, and we live in a media saturated world. 

Let’s say a millennial man is scrolling Instagram, sees an ad for a new, trending microbrand watch, clicks, and then buys. Score one for the ad, right? Maybe it’s not so simple.

What if that same man was already shopping for buzzy boutique watches? He had this new microbrand on his shortlist after hearing about it on a podcast, reading Reddit reviews, and seeing it flashed on the wrist of his favorite TikTok creator. The Instagram ad was certainly opportune and helped close the deal. But how much credit does it deserve?

Let’s imagine another guy is in the same situation, hyped on the watch brand, but never sees the Instagram ad.  Yet he buys one, too. How important was that campaign in truly moving the needle on sales? Especially since the first man seemed close to buying it anyway.

How do we close the gap and better isolate outcomes to a campaign?

The answer to this question is: incrementality.

Understanding Incrementality 

According to the Ad Age Encyclopedia, “Incrementality in advertising measures the true impact of a marketing effort by determining what effect, if any, it had beyond what would have happened anyway. It isolates the lift in behaviors—such as purchases, clicks or sign-ups—that can be directly attributed to a specific campaign or tactic.”

In the watch example, the second guy, who never saw the Instagram ad, is part of our control – the baseline of “what would have happened anyway.” We then use these actions (or inaction) that are not touched by the campaign to compare against those that were. We see “lift” when we measure comparatively more outcome with consumers who were exposed to the campaign.

The challenge, of course, is establishing the baseline. 

Brand lift to the rescue.

How Brand Lift Measures Incrementality

As you might guess from the “lift” part of its name, incrementality is built into brand lift’s DNA. Establishing a baseline is a necessary part of the method.

So, how does brand lift establish a baseline?

Brand lift measures how an ad campaign affects, and changes, perception of a brand by using surveys. We ask a person what brands they can recall from a certain category – say, watches – what they think about the brands, and how likely they are to consider or purchase any. 

With this method we use two cells of survey respondents. The “control” cell serves as the baseline, and is made up of consumers who have not been exposed to the ad campaign. This provides us with a snapshot of a brand’s or product’s existing health in the marketplace – how well consumers know it and think about it, absent exposure to the campaign in question.

The “exposed” cell consists of respondents who were exposed to the campaign, and represents the outcomes associated with hearing or seeing the ad. 

Respondents in both cells answer identical questions for brand metrics so we can compare their results apples-to-apples. Any difference between the two cells is our “lift” or “incrementality.”

Case Study: Incremental Lift for a Home Internet Brand

To make this more vivid, let’s look at brand lift results from a recent YouTube podcast ad campaign for a well-known home internet service. We know the brand is well-known because a strong majority (72%) of the control cell – those who did not see the campaign – said they are very or extremely familiar with it. Even with a high baseline, the campaign still  moved the needle some, with 76% of the exposed cell – who saw the ad – indicating familiarity.

Judged on its own, seeing that three-quarters of the respondents exposed to the ad are familiar seems really good. Comparing that to the 72% baseline puts matters in sharp perspective.

However, we also know it’s challenging for well-known brands with strong brand health in the marketplace to achieve big lifts in top-funnel measures, like feasibility, in general. So let’s travel down the funnel. 

A little more than half (52%) of respondents who saw the brand’s ad said they were likely to consider it. While that’s a slight majority, judged in isolation a buyer or CMO might be unclear on how good that result is.

When we compare it to the control baseline, things snap right into focus. Only 40% of the control respondents indicated they would consider the brand, a 12 point difference. That result was also statistically significant, meaning we are confident that if we repeated this survey 100 more times, we would see a lift in consideration at least 90 times. 

Put another way, the brand lift study measured a significant incremental lift in consideration for the home internet brand. Without seeing this specific ad campaign, a baseline 40% of consumers say they’d consider the brand. By comparison a larger proportion of consumers – 52% – who saw an ad say they would consider the service.

Adding brand lift to your ad effectiveness toolbox  means incrementality isn’t hard to measure, especially when combined with attribution.