How to Choose the Right Sample Size for Surveys and Experiments

Sample size decides how much trust you can place in survey or experiment results. Too small, and the result may swing wildly from sample to sample. Too large, and you may spend more time or money than the decision requires.

The Three Inputs That Matter Most

The usual sample size calculation depends on confidence level, margin of error, and expected proportion. A 95% confidence level is common because it balances reliability and practicality. A smaller margin of error requires a larger sample because you are asking for more precision.

The expected proportion matters when you are estimating a percentage, such as the share of users who prefer a feature. When you are unsure, 50% is often used because it creates the most conservative sample size.

Why Bigger Is Not Always Better

A larger sample can reduce uncertainty, but the benefit slows down. Moving from 100 to 400 responses is a big improvement. Moving from 10,000 to 10,300 responses usually adds very little practical value. Good sampling design is often more important than raw sample count.

Common Mistakes

• Using a sample that is large but biased.
• Ignoring non-response bias in surveys.
• Choosing a margin of error after seeing the results.
• Confusing statistical precision with business importance.

Calculate It Yourself

Use the Sample Size Mean Calculator before collecting numeric data, then review related tools from the Statistics calculators hub for the rest of your survey planning workflow.