Rex says
The fast lane for the math you almost remember from school. Type the numbers, get the answer, move on with your day.
Try a scenario
Click to load — tweak from there.Inputs
Result
People to invite per group
15,600
Completed responses needed per group
1,248
Total invites needed (both groups)
31,200
Target metric after lift
30.0%
Psst — share this and help Rex grow
One click, a permanent link with your numbers baked in.
How to use this
- 1Enter baseline metric (e.g. aided awareness) (%).
- 2Enter target lift in percentage points (pts).
- 3Enter confidence level.
- 4Enter expected survey response rate (%).
- 5Read your people to invite per group on the right — it updates as you type.
- 6Hit Share to keep the scenario or send it to someone.
About this calculator
Brand lift studies survey an exposed group and a control (unexposed) group on awareness, favorability, or purchase intent, then compare the two to measure the campaign's effect on brand metrics rather than clicks or conversions. Because survey response rates are usually low and the metric being measured (like 'aided awareness') often starts from a high baseline with a smaller expected lift than direct-response metrics, brand lift studies typically need larger respondent pools per cell than a standard conversion A/B test to detect a real difference confidently. This calculator applies the same two-proportion sample size formula used for conversion tests to survey metrics, then grosses up for expected survey response rate to tell you how many people you need to invite into each arm, not just how many need to complete the survey. A study measuring a baseline 25% aided awareness with a target 5-point lift to 30% needs meaningfully more completed responses per cell than a direct-response test detecting a 20% relative lift on a 3% baseline conversion rate, because the absolute gap being detected is proportionally smaller relative to the baseline variance.
Worked example
Using the values the calculator loads with:
Inputs
- Baseline metric (e.g. aided awareness): 25 %
- Target lift in percentage points: 5 pts
- Confidence level: 95%
- Expected survey response rate: 8 %
Results
- People to invite per group: 15,600
- Completed responses needed per group: 1,248
- Total invites needed (both groups): 31,200
- Target metric after lift: 30.0%
What each field means
Inputs
- Baseline metric (e.g. aided awareness) (%)
- The baseline metric (e.g. aided awareness) used in the calculation, measured in %. Starts at 25 % so you have a working example on load. Accepted range: 1–99 %.
- Target lift in percentage points (pts)
- The target lift in percentage points used in the calculation, measured in pts. Starts at 5 pts so you have a working example on load. Accepted range: 0.5–50 pts.
- Confidence level
- Pick the option that matches your situation — the maths changes per option. Choices: 90%, 95%.
- Expected survey response rate (%)
- The expected survey response rate used in the calculation, measured in %. Starts at 8 % so you have a working example on load. Accepted range: 0.5–100 %.
Results
- People to invite per group
- Returned as a whole number and shown as the headline result. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
- Completed responses needed per group
- Returned as a whole number. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
- Total invites needed (both groups)
- Returned as a whole number. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
- Target metric after lift
- Returned as a percentage. It recalculates instantly whenever you change an input, so you can compare scenarios without reloading.
FAQ
Why do brand lift studies need so many more respondents than a conversion test?
Conversion tests often measure larger relative lifts (10-30%) on metrics with plenty of raw volume. Brand surveys typically target smaller absolute point lifts (3-7 points) on a metric that already has real variance around a higher baseline, which the sample size formula punishes heavily since it divides by the squared gap.
What response rate should I plan for?
In-platform intercept surveys (like Meta or YouTube brand lift studies) built into the ad experience often see 3-10% response rates. Email or panel-based surveys sent after the fact usually run lower, 1-3%, unless there's a strong incentive attached.
Can I use a smaller sample if I only care about directional results?
You can run smaller and treat results as directional, but be honest that a study underpowered for its target lift will frequently show 'no significant difference' even when a real lift exists, which risks killing a campaign that's actually working.
Should the control group see a different ad or no ad?
True control should see either no ad (a public service announcement / ghost ad in-platform) or an unrelated ad, never the same brand's other creative, otherwise you're measuring creative preference rather than the campaign's incremental lift on brand metrics.
Accuracy and limitations
- Results are rounded for display; the underlying calculation keeps full precision.
- Very large or very small inputs may hit floating-point limits in the browser.
- Inputs outside the accepted range are clamped rather than rejected.
Related tools
Cite this calculator
Writing about this topic? Grab a citation — every link helps keep these tools free.
RevenueLab. (2026). Brand Lift Survey Sample Size Calculator. Retrieved from https://www.revenuelab.fyi/toolbox/brand-lift-survey-sample-size
<p>Source: <a href="https://www.revenuelab.fyi/toolbox/brand-lift-survey-sample-size" target="_blank" rel="noopener">Brand Lift Survey Sample Size Calculator — RevenueLab</a> (2026).</p>
Source: [Brand Lift Survey Sample Size Calculator — RevenueLab](https://www.revenuelab.fyi/toolbox/brand-lift-survey-sample-size) (2026).