Brand Lift Calculator
Measure brand lift from a control versus exposed group study, calculate absolute and relative lift, and check whether the result is statistically meaningful.
📈 Marketing📐 Absolute lift = exposed rate − control rate; relative lift = absolute / control rate💼 Business
Control group size
Control group positive responses
Exposed group size
Exposed group positive responses
Media spend for the campaign
Please enter valid values.
Formula & Reference
| Variable | Symbol | Formula | Units |
|---|---|---|---|
| Brand Lift Calculator | — | Absolute lift = exposed rate − control rate; relative lift = absolute / control rate | percentage points |
Step-by-Step Examples
Example 1
Significant Lift
Control 288 of 2,400; exposed 416 of 2,600; spend 85,000.
- Control rate = 12.00%, exposed rate = 16.00%
- Absolute lift = 4.00 percentage points
- Relative lift = 33.3%
- Z = 4.02 — significant at the 99% level
- Cost per lifted response = 85,000 / 104 = 817
✓ 4.00 points, 33.3% relative, significant
Example 2
Not Significant
Control 240 of 2,000; exposed 260 of 2,000.
- Control 12.00%, exposed 13.00%
- Absolute lift = 1.00 point, relative 8.3%
- Z ≈ 0.97 — not significant at conventional levels
- The observed difference could plausibly be noise
✓ 1.00 point — not significant
Example 3
Small Base Warning
Control 12 of 100; exposed 20 of 100.
- Absolute lift = 8.00 points, relative 66.7%
- The relative lift looks dramatic
- Z ≈ 1.62 — below the 95% threshold despite the large apparent effect
✓ Large apparent lift, insufficient sample
Real-World Applications
Upper Funnel Measurement
Brand lift studies measure awareness and consideration effects that click-based attribution cannot capture.
Controlled Comparison
Randomised control and exposed groups isolate campaign effect from background trends.
Statistical Discipline
Testing significance prevents reading noise as effect, which is common with small survey samples.
Cost Per Lifted User
Dividing spend by incremental positive responses gives a comparable efficiency measure across campaigns.
Common Mistakes to Avoid
⚠️
Reporting relative lift without absolute
A 66% relative lift sounds impressive but may represent a move from 3% to 5% on a small base. Both figures are needed for honest reporting.
⚠️
Ignoring statistical significance
With small samples, apparent differences are frequently noise. Testing significance is essential before acting on the result.
⚠️
Comparing non-equivalent groups
Control and exposed groups must be randomly assigned. Self-selected or differently-sourced groups introduce bias that invalidates the comparison.
Frequently Asked Questions
What is brand lift? ▾
The increase in a brand metric such as awareness, recall, or consideration among an exposed group compared with a matched control group.
What is the difference between absolute and relative lift? ▾
Absolute lift is the difference in percentage points; relative lift expresses that difference as a percentage of the control rate. Relative figures look much larger.
How do I know if lift is significant? ▾
A two-proportion z-test compares the groups. A z-statistic above 1.96 indicates significance at the conventional 95% confidence level.
What sample size do I need? ▾
It depends on expected effect size and baseline rate. Detecting small lifts on low baseline rates can require several thousand respondents per group.
Why measure brand lift at all? ▾
Because upper-funnel effects on awareness and consideration do not show up in click or conversion tracking, yet influence long-term demand.