Solution Finder
What is your customer trying to achieve? Pick the problem closest to what they actually said — this translates it into the right methodology, data requirements, and Lab.
You said: “Measure whether advertising changed brand perception”
Brand Lift Measurement
MeasurementWhy this fits
Brand Lift Measurement directly compares perception metrics between an exposed and control group.
What 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 required
- 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.
Implementation approaches
Risks & limitations
Risks
- Selection bias if exposed and control groups differ systematically before the campaign
- Low statistical power if sample sizes are too small to detect real lift
Limitations
- Survey-based measurement reflects stated perception, not guaranteed behaviour change
- Platform-native brand lift is not independently verified and may have different exposure logic