Calculator Inputs
Enter survey and business inputs below. Results will appear above this form after submission.
Example Data Table
| Scenario | Exposed Group | Control Group | Outcome |
|---|---|---|---|
| Brand awareness survey | 2,500 respondents, 1,225 positive | 2,400 respondents, 960 positive | Exposed rate 49.00%, control rate 40.00%, lift 9.00 points |
| Post-campaign conversions | 4.8% conversion rate | 3.9% conversion rate | 0.9 point incremental conversion lift |
| Commercial estimate | Audience reached 500,000 | Average order value $72 | Supports projected revenue and ROI calculations |
Formula Used
1) Exposed and Control Rates
Exposed Rate = Exposed Positive Responses / Exposed Sample Size
Control Rate = Control Positive Responses / Control Sample Size
2) Absolute Brand Lift
Absolute Lift = Exposed Rate - Control Rate
This is shown in percentage points.
3) Relative Lift
Relative Lift % = ((Exposed Rate - Control Rate) / Control Rate) × 100
4) Statistical Significance
Pooled Rate = (Exposed Positives + Control Positives) / (Total Sample)
Standard Error = √[Pooled Rate × (1 - Pooled Rate) × (1/Exposed Sample + 1/Control Sample)]
Z Score = (Exposed Rate - Control Rate) / Standard Error
5) Confidence Interval
CI = Absolute Lift ± Z Critical × Unpooled Standard Error
6) Business Impact
Lifted Users = Audience Reached × Absolute Lift
Incremental Conversions = Audience Reached × (Exposed Conversion Rate - Control Conversion Rate)
Incremental Revenue = Incremental Conversions × Average Order Value
Gross Profit = Incremental Revenue × Gross Margin
ROI % = ((Gross Profit - Campaign Cost) / Campaign Cost) × 100
How to Use This Calculator
- Choose the brand metric you want to test, such as awareness, consideration, favorability, or purchase intent.
- Enter the exposed and control sample sizes from your survey or experiment.
- Enter how many positive responses each group produced for the selected brand metric.
- Select a confidence level to control the strictness of the significance test.
- Add audience reached, campaign cost, conversion rates, order value, and gross margin for commercial estimates.
- Submit the form to see rate comparison, absolute lift, relative lift, p-value, interval, and revenue effect.
- Use the export buttons to save results as CSV or PDF for reporting, planning, or stakeholder review.
Frequently Asked Questions
1) What does brand lift measure?
Brand lift measures how much a campaign improves a target metric in the exposed group compared with a control group. Common metrics include awareness, consideration, recall, favorability, and purchase intent.
2) Why is a control group important?
A control group estimates the baseline response without campaign exposure. Comparing exposed and control groups helps isolate the campaign effect from seasonality, prior demand, or unrelated market changes.
3) What is the difference between absolute and relative lift?
Absolute lift is the raw percentage point difference between groups. Relative lift shows that difference as a percentage of the control rate, helping you compare campaigns with different baselines.
4) What does statistical significance mean here?
Statistical significance tests whether the measured difference is likely due to the campaign rather than random sampling variation. A small p-value suggests stronger evidence that the lift is real.
5) How are lifted users estimated?
Lifted users are estimated by multiplying the campaign reach by the absolute lift rate. This gives an approximate count of additional people influenced on the measured brand metric.
6) Can this calculator estimate revenue impact?
Yes. When you provide conversion rates, average order value, and gross margin, the calculator estimates incremental conversions, revenue, gross profit, and ROI from the measured campaign effect.
7) What happens if sample sizes are too small?
Small samples create wider confidence intervals and less stable significance results. You may still see lift, but uncertainty will be higher and decisions should be made more carefully.
8) Can I use this for different brand metrics?
Yes. The calculator works for any binary survey outcome, such as aware versus unaware, favorable versus unfavorable, or intent versus no intent, as long as both groups are comparable.