Examples

A/B Test Sample Size Examples

Examples make A/B test planning easier because they show how assumptions change the traffic requirement. This page focuses on intuition rather than formal proof.

Example: modest baseline, modest lift

Imagine a page with a 10% baseline conversion rate and a target lift of 2 percentage points. That kind of setup often produces a manageable but still meaningful sample requirement.

It is a good example because it sits in the range many growth and product teams actually work with.

Example: small baseline, tiny lift

Now imagine a 2% baseline and a 0.3 percentage-point lift. The required sample rises quickly because the effect is small and the baseline is relatively low.

This is why very small target effects can make experiments impractical for low-traffic experiences.

The practical lesson

Examples like these show that the key question is not just how much traffic you have. It is whether the effect you care about is large enough to detect within the traffic you can realistically collect.

That is what makes sample size planning a strategic filter for experimentation.

How to learn from examples without copying them

Examples are most useful for building intuition about how assumptions interact. They show how baseline, effect size, and power can move the traffic requirement much more than people expect.

What examples should not do is replace your own planning inputs. Even a very similar-looking test can need a different sample if the audience, metric, or baseline behaves differently.

  • Use examples to understand direction, not to copy a sample target
  • Compare your own baseline and MDE against the example setup
  • Treat example numbers as illustrative rather than prescriptive
  • Run your own scenario in the calculator before launching

Related pages for A/B Test Sample Size Examples

Frequently Asked Questions

What will I learn on this page?
Examples make A/B test planning easier because they show how assumptions change the traffic requirement. This page focuses on intuition rather than formal proof.
Who is this A/B testing guide for?
This guide is for product teams, growth marketers, analysts, and anyone planning experiments who wants to make better decisions about effect size, traffic, and test design.
What should I do after reading this page?
Use the explanation here to choose realistic assumptions, then move to the calculator or related pages to estimate the traffic needed for your experiment.
Why are examples useful if they are not exact matches to my test?
They help you build intuition about how baseline, MDE, and power move the traffic requirement. That makes it easier to spot whether your own test assumptions are conservative, aggressive, or unrealistic.
What should I compare between an example and my own test?
Compare the baseline conversion rate, the target lift, the confidence and power settings, and the amount of traffic available. Those factors usually matter much more than surface similarities between experiments.