Recommended: For a more comprehensive tool that visualizes how p-values and confidence intervals evolve over time with daily data input, check out our P-Value and Confidence Interval Over Time Calculator.

← Free tools

Confidence Interval Calculator (Snapshot)

Calculator Configuration

/
/
95%
90%99%

Recommended: For a more comprehensive tool that visualizes how p-values and confidence intervals evolve over time with daily data input, check out our P-Value and Confidence Interval Over Time Calculator.

How it works

Watch a demo of the Confidence Interval Calculator ⤴

What is this?

This tool calculates a confidence interval (CI) for the difference in conversion rates between two groups (typically a control and a variant) using just the final totals for each.

Why use it?

P-values only tell you whether a result is statistically significant. Confidence intervals tell you how big the difference might be, and how confident you can be about that range.

Use this tool to:

  • Quantify the uncertainty around your observed uplift
  • See if the interval includes zero (no effect)
  • Determine whether the possible outcomes are meaningfully positive or negative

How it works

You enter:

  • Total conversions and visitors for the control group
  • Total conversions and visitors for the variant group
  • Desired confidence level (e.g. 95%)

The calculator:

  • Computes conversion rates
  • Calculates the absolute and relative lift
  • Generates a confidence interval (CI) around the difference using the Wald method

Formula used

The CI is calculated on the difference of proportions:

CI=(p^1p^2)±Zα/2SECI = (\hat{p}_1 - \hat{p}_2) \pm Z_{\alpha/2} \cdot SE

Where:

  • p^1\hat{p}_1 and p^2\hat{p}_2 are the observed conversion rates
  • SESE is the standard error of the difference in proportions
  • Zα/2Z_{\alpha/2} is the z-score corresponding to your confidence level (e.g. 1.96 for 95%)

If the CI includes 0, the result is not statistically significant at the selected confidence level.

Use this to get a range of plausible outcomes, not just a binary yes/no.

Work with us

Want a second set of eyes on your experiments?

Book a free 15-minute call. Bring a test you're running or planning. We'll look at the design, the instrumentation, or the results together.

Book a 15-minute call