How Big a Test?

We use a widely accepted statistical formula to calculate sample size.

This calculator can be used for traditional A/B (Split Run) testing as well as more statistically complex scientific testing.

Generally, the smaller the change you want to see and/or the lower your response rate, the larger the sample size you need.

Sample-Size Calculator

Based on the type of test and the confidence level you require, how many contacts does your test list need to return results for which you can have a high degree of confidence?

Test Type

Split Run Test Set-Up
Parameter Value Explanation
Number Variables Use of value of 1 if you have data from a previous mailing to use as your baseline.
Values Per Variable A/B test, A/B/C test, etc. The control value should be counted as one.
Typical Response Rate Without a test, what response rate would you expect?
Significance Threshold The lower the percentage, the bigger the required list.

Scientific Test Set-Up
Parameter Value Explanation
Typical Response Rate If you weren't doing a test, what would you would expect to achieve?
Significance Threshold The lower the percentage, the bigger the required list.


Our privacy policy is simple: Your data is private. We don't use it, analyze it, share it, aggregate it or otherwise access it for any purpose other than to run this utility.

Our tools are designed to work with partial records. For the name modification/evaluation tools, feel free to just share the firstname and a unique identifier with us, then merge the results back into your primary list. That way, you won't have to share your email or physical address list with us.

Whenever sharing your list with a utility or vendor like us, always seed your list with addresses that, in the event they received communication, would demonstrate unauthorized access to your data.