How many responses you need for a statistically valid survey.
Required sample size
385
Assumes large population (no adjustment)
Estimated outreach (based on response rate)
Before you run a survey, you need to know how many responses to collect. Too few and your results are noise. This calculator uses your population size, desired confidence level, and acceptable margin of error to give you the minimum sample size for statistically meaningful results.
Z is the z-score for your confidence level (1.96 for 95%). p is the response distribution (use 0.5 for maximum variability). E is your margin of error as a decimal. The formula is then adjusted for your population size when it is small.
Typical confidence level
95%
Typical margin of error
5%
Min responses for NPS
100+
Produktly gives you tours, checklists, NPS, surveys, feedback widgets, roadmaps, and changelogs in one tool. Starting at $19/mo.
95% is standard for business decisions. 90% is acceptable for quick directional reads. 99% is overkill outside academic or medical research.
5% is standard. 3% gives more precision but requires far more responses. 10% is too loose for most decisions.
This calculator is for surveys (estimating a single value). A/B tests need a different calculation that accounts for the difference between two variants and statistical power. Use a dedicated A/B test calculator for that.