top of page
Search

Brand Lift Study: What It Measures and How to Run One

Writer: Vain.
Vain.
1 day ago
9 min read

Participants answering an in-app brand survey

A brand lift study is a survey-based experiment that shows whether your advertising changed what people think of your brand. It compares people who could see your campaign with a matched group who could not, asks both the same short questions, and treats the gap between their answers as the lift. If your goal is awareness, recall, favorability or consideration, this is the cleanest way to prove it. And because the result depends so heavily on the creative being tested, the work you put in before launch matters as much as the survey itself.

 

This guide is written for brand managers, marketers and creative teams in New York City and beyond who commission video, audio and social work and need to know whether it moved anyone.



TL;DR

 

  • A brand lift study splits an audience into an exposed group and a control group, surveys both, and reports the difference as lift.

  • The common metrics are awareness, ad recall, familiarity, favorability, consideration and message association.

  • Platform studies are efficient but often carry budget minimums and only measure what happens inside that platform.

  • Brand lift answers “did perception change?” Conversion lift answers “did people buy more?” Pick the one that matches your campaign goal.

  • Results are only as strong as the creative and the one or two KPIs you design it around.

 

Table of Contents

 

 

What a brand lift study actually is

 

Think of it as a controlled experiment for attention and attitude. Instead of guessing whether a video, a podcast read or an animated spot made people more aware of you, you measure it.

 

The platforms that run these studies describe the same core design. Snapchat’s help center explains that the audience is randomized into a test group eligible to see your ads and a control group that is not. Both groups answer a short in-app survey while the campaign is live, and the difference in responses is the lift. Amazon Ads describes a version that surveys its shopper panel on awareness, attitudes, preferences, favorability, intent and ad recall, with survey questions built from templates and results available on demand.

 

The details differ by platform, but the logic is constant: same questions, two groups, one difference.


What a brand lift study actually is

How the exposed-versus-control comparison works

 

The reason this works is that the control group stands in for the people who would have answered the survey anyway. Some of your audience already knew your brand. Some would have liked it regardless. The control group tells you how large that baseline is, so the lift reflects what the campaign added.

 

Here is the arithmetic, using a worked example from Signal Hill Insights’ explanation of audio brand lift. If 45% of exposed listeners say they intend to purchase and 30% of the matched control group says the same, the lift is 15 percentage points. You can also express it as relative lift, the difference divided by the baseline. In that example, 15 divided by 30 is 50% relative lift.


Exposed and control purchase intent comparison

Be clear about which version appears in your report. A “15-point lift” and a “50% lift” describe the same result and sound very different in a deck.

 

Statistical confidence

 

A gap between two groups could be noise, so studies report significance. A 2018 American Statistical Association paper on brand lift of online ads describes lift as the difference between treatment and baseline response rates, with significance determined by one-sided tests at the 0.1 level, which produces 80% confidence intervals. That is a looser threshold than many marketers assume. Other providers use 90% or 95%, so always check which level your study applies before you compare results across vendors or campaigns.

 

Sample size drives all of this. Small exposed or control groups produce wide intervals, and a campaign with modest reach may never reach significance even if the effect is real.

 

The metrics a brand lift study can measure

 

Most studies draw from the same menu. Choose the ones your creative was built to move.

 

Metric

The question it answers

Best suited to

Awareness

Have you heard of this brand?

New brands, new markets

Ad recall

Do you remember seeing an ad from this brand?

Video, audio and social campaigns

Familiarity

How well do you know this brand?

Challenger brands

Favorability

How do you feel about this brand?

Reputation and culture-led work

Consideration or purchase intent

Would you consider buying from this brand?

Launches and seasonal pushes

Message association

Do you link the brand with a specific idea?

Positioning and storytelling campaigns

This six-metric framing follows Dynata’s overview of brand lift metrics. For a plain-language look at how brand and media metrics fit together, this guide to earned media metrics and how to measure them is a useful companion.

 

Focus on one or two KPIs

 

It is tempting to measure everything. Don’t expect to move everything. Signal Hill’s Paul Riismandel points out that a campaign can lift awareness significantly without lifting recommendation intent, and advises focusing creative on one or two KPIs and judging results against those.

 

That is a creative brief decision, not a survey decision. A spot designed to make a name memorable will be built differently from one designed to change how people feel about a brand. Decide which job the work has before production starts, and the study will have something to measure.

 

How to run a brand lift study, step by step

 

1. Define the single question the campaign must answer

 

Write it as a sentence: “After this campaign, more people in our target audience should be able to recall our brand.” If you cannot finish that sentence, you are not ready to measure.

 

This is also where creative strategy and measurement meet. When we build a strategy with a brand, we start from the outcome the work has to deliver, so the video, audio or animation is shaped around it from the first brief.

 

2. Choose the right KPIs

 

Pick one or two from the table above. Match them to the funnel stage: awareness and recall for new audiences, favorability and association for brand repositioning, consideration for launches.

 

3. Pick where the study will run

 

You have three broad routes:

 

  • A platform’s built-in study, run by the same network that serves the ads. These are convenient and integrate with the campaign, but they measure only that platform’s impact.

  • An independent survey panel, which can follow exposure across several channels.

  • A study on a specialist channel, such as audio or podcasts, where exposure is tracked differently from video.

 

Many platform studies have minimum budgets, and those minimums change. Before planning, check the platform’s current requirements on its own help page rather than relying on a figure you saw in an article, including this one. Quantcast’s Brand Lift Live documentation, for instance, notes that survey budget is additional to campaign spend and is split between exposed and control delivery to maximize statistical significance. Budget for the survey, not only the ads.

 

4. Build clean survey questions

 

Keep them short and neutral. A one-question survey is common, and it limits drop-off. Avoid leading wording, and ask the same question of both groups in the same order. Platform templates, like the pre-populated ones in Amazon’s tool, help you avoid the most common mistakes.

 

5. Lock the creative before launch

 

The study tests what you put in market. Changing the edit, the cut-down lengths or the audio read midway makes results hard to interpret. Finalize hero assets, variations and placements first, and let the campaign run long enough to collect a meaningful sample.

 

6. Read the results against your KPI

 

Look at the lift, the confidence level and the sample sizes together. A positive lift that is not statistically significant is a hint, not a result. A significant lift on a metric you did not design for is a bonus, not a victory lap.

 

Brand lift for audio and podcasts

 

Audio is where many brands struggle to prove value, because listeners rarely click. The brand lift method fits well here: a matched panel of listeners who did not hear the campaign answers the same questions as those who did, and the control scores are subtracted from the exposed scores, as in the purchase-intent example above.

 

If you are planning podcast work, budget for production, promotion and measurement together. Our guide to podcast budgets that cover promotion and KPIs for NYC brands walks through how to allocate spend, and our 90-day branded podcast plan shows how to sequence it. For a sense of how audio platforms and culture-led channels reach listeners, take a look at this page on promoting audio and music content.

 

Brand lift versus conversion lift

 

The two measure different things, and confusing them leads to disappointing reports.

 

Brand lift

Conversion (sales) lift

Question

Did perception change?

Did behavior change?

Method

Survey of exposed vs. control

Observed purchases or actions, exposed vs. control

Typical metrics

Awareness, recall, favorability, intent

Sales, sign-ups, store visits

Best for

Upper- and mid-funnel creative

Performance and lower-funnel campaigns

Time to signal

During the campaign

Often needs a longer purchase window

The explainer from afaqs! makes the same distinction. A brand lift study asks what people would consider buying, for example 36% consideration among the exposed versus 30% among the control for a six-point lift. Sales lift asks whether exposed people actually bought more. Incrementality is the broader idea: the extra business that would not have happened without the ads.

 

A simple rule: if the campaign is meant to make people remember and prefer you, run a brand lift study. If it is meant to drive a transaction you can count, run a conversion lift test. For brands with both goals, run both, and read each against its own question. Our case-study write-up on creative campaigns and a 5% sales lift shows how a creative brief can be written with a sales outcome in mind.

 

Know the limits of platform-native studies

 

Platform studies measure incrementality inside the platform running them. Measured.com’s Terence Einhorn notes that this leaves cross-channel effects invisible, and suggests using lift tests to establish causal ground truth, then feeding the findings into marketing mix modeling for the always-on, cross-channel view.

 

In practice, that means a lift study is a snapshot of one campaign, not a verdict on your whole media plan. To see how channels work together, read our explainer on media mix and our 90-day marketing analytics playbook.

 

Make the creative worth measuring

 

A lift study can only report what the work earned. Three habits raise the odds of a positive result:

 

  1. Build around one idea. Campaigns that try to say five things tend to be remembered for none.

  2. Design for recall from the first seconds. Brand cues, sound and a distinctive visual language help viewers connect the work to you.

  3. Match the audience to the story. Layered segmentation keeps the message relevant. See our audience segmentation examples for ways to structure it.

 

Culture-led work often performs well on favorability and association, because it gives people a reason to care. Our roundup of cultural marketing examples shows how brands have done it.

 

Where we come in

 

We are a creative media and technology company in New York City. Our video, audio, animation and strategy services are built on a curated network of collaborators, and we work with brands, artists and organizations to shape content and campaigns that make the ordinary extraordinary. If you are planning a campaign you intend to measure, we can help you define the outcome, shape the strategy and produce the creative that the study will test.


Vainnewyork

Explore our video, audio, animation and strategy services and tell us what your campaign needs to move.

 

Our perspective

 

Every great production is a tapestry of talent: writers, directors, editors, sound designers and animators pulling toward one idea. A brand lift study is the moment that idea meets an audience, and it is the fairest test a campaign can get.   We believe measurement should start in the creative brief, not after the wrap. When a team agrees up front on the one or two things a piece of work must change, every choice from the script to the sound mix becomes sharper, and the lift study becomes a confirmation rather than a hope.   That is why we pair strategy with production. A curated network of collaborators lets us bring the right craft to each project, whether it is video, audio or animation, so the work is intentional from first frame to final cut.   Vain.

 

FAQ

 

What is a brand lift study?

 

It is a survey-based experiment that compares people who were eligible to see your ads with a control group who were not. Both answer the same questions about awareness, recall, favorability or intent. The difference between their answers is the lift your campaign produced.

 

What is brand lift measurement used for?

 

It shows whether a campaign changed perception, which clicks and views cannot tell you. Marketers use it to compare creative versions, justify upper-funnel spend and decide which messages to carry into the next campaign.

 

How much does a brand lift study cost?

 

It depends on the provider and the platform. Some platforms offer the study as part of an eligible campaign, but they often set minimum ad spend, and those minimums change. Independent studies are priced separately. Check the platform’s current requirements page and add survey budget to your plan.

 

What is the difference between brand lift and conversion lift?

 

Brand lift measures changes in attitudes through surveys, such as awareness or consideration. Conversion lift measures changes in actions, such as purchases or sign-ups. Use brand lift for awareness and perception goals, and conversion lift for sales goals.

 

How long should a study run?

 

Long enough to gather a sample large enough for statistically significant results. That depends on your reach and the platform’s rules, so plan the flight length and budget together and keep the creative unchanged while the study runs.

 

Sources

 

Recommended

 

 
 
 

Comments


bottom of page