Customer question
“Did advertising change how people think about our brand?”
What this solution does
You ask people who saw the ads, and people who didn't, the same questions about the brand — the gap between their answers is the campaign's effect on perception.
Specifically, it answers: Did a campaign move perception metrics like awareness, consideration, or purchase intent, and by how much, compared to people who weren't exposed to it?
How it works
An exposed group (saw the ads) and a control group (didn't) are surveyed on the same brand perception questions; the difference in responses between groups is the estimated lift attributable to the campaign.
Data requirements
- A defined exposed population (saw the campaign)
- A comparable control population (did not see the campaign)
- Survey responses on brand perception metrics from both groups
- Sufficient sample size in both groups for statistical reliability
Discovery questions
How will exposed and control groups be defined and separated?
Poor separation between groups (e.g. control group partially exposed) biases the lift estimate toward zero.
What sample size is achievable in each group?
Small samples produce wide confidence intervals, making it hard to distinguish real lift from noise.
Is this measured via an independent research panel or a platform-native tool?
Platform-native brand lift tools use the platform's own exposed/control logic, which can differ from independent methodology and carries different bias considerations.
Build an experiment
Generate synthetic data
Data readiness
Run model / simulation
Results
Business ⟷ Technical interpretation
Awareness moved from controlAwareness% (control) to exposedAwareness% (exposed) — a awarenessLift percentage point lift — with purchase intent showing a smaller but still positive intentLiftpp lift.
Shown as a template — the highlighted tokens are filled in once the model has been run.
What would you do next?
- Extend the study if confidence intervals are too wide to act on confidently
- Compare independent brand lift results against any platform-native brand lift reporting for the same campaign
Solution & vendor landscape
Implementation approaches
Related concepts