📐
How to use the calculator
  1. Enter clicks and conversions from your advertising or analytics report.
  2. Choose a confidence level: 95% corresponds to α = 0.05.
  3. Read the Wilson confidence interval for the underlying conversion rate.
  4. Use the lower bound as a conservative CR input for CPC planning.
Confidence level — select a card
α = 0.10
Confidence 90%
In repeated sampling, about 90% of intervals cover the underlying rate. Narrower interval.
α = 0.05
Confidence 95%
In repeated sampling, about 95% of intervals cover the underlying rate. A common choice.
α = 0.01
Confidence 99%
In repeated sampling, about 99% of intervals cover the underlying rate. Wider interval.
Lower bound
—
—
Observed CR
—
—
Upper bound
—
—
— ↑ Observed CR —
Enter clicks and conversions to see the result.

What a conversion rate confidence interval means

The observed conversion rate is conversions divided by clicks. With a small sample it can differ greatly from the underlying rate. A confidence interval expresses that uncertainty.

Wilson score interval

This calculator uses the Wilson score method, which behaves better than the basic normal approximation for small samples and rates near 0% or 100%.

A 95% confidence procedure would produce intervals containing the fixed underlying rate in about 95% of repeated samples under the model. It does not assign a 95% probability to that fixed rate after observing this sample. The selected α is a significance level, not an observed p-value.

Using the result for PPC planning

A lower bound can be used as a conservative input when estimating maximum CPC from conversion value or target CPA. It is not a guaranteed future rate: audience changes, seasonality and tracking errors still matter. This binomial calculation assumes independent clicks and at most one conversion per click.

For 200 clicks and 3 conversions the observed CR is 1.50%; the 95% Wilson interval is approximately 0.51%–4.32%. More observations generally narrow the interval. There is no universal conversion count that makes every campaign reliable.

Compare like-for-like traffic and use a stable conversion definition. Microconversions can be useful diagnostics but should not be treated as purchases with the same value.