Enter your population size, confidence level, and margin of error to calculate how many responses you need for statistically valid survey results.
Calculate the minimum number of survey responses needed for statistically valid results.
Total number of people you could survey
How certain you want to be (95% is standard)
Acceptable error range (5% is standard)
Quick Reference: Sample Sizes at 95% Confidence
| Population | ±3% | ±5% | ±10% |
|---|---|---|---|
| 100 | 92 | 80 | 49 |
| 500 | 341 | 217 | 81 |
| 1,000 | 516 | 278 | 88 |
| 5,000 | 879 | 357 | 94 |
| 10,000 | 964 | 370 | 95 |
| 50,000 | 1,045 | 381 | 96 |
| 100,000+ | 1,067 | 383 | 96 |
These values assume: 95% confidence, p = 0.5, and finite population adjustment.
Notice an important pattern: once your population exceeds 50,000, sample size barely increases. A population of 50,000 needs 381 responses at ±5%, while 100,000+ needs only 383. This is why national polls work with just 1,000-1,500 respondents.
How the Calculator Works
This calculator uses Cochran’s formula, the standard method for determining survey sample size. It works in two steps:
- Calculate the initial sample size assuming an infinite population, using your confidence level and margin of error
- Apply the finite population correction to adjust for your actual population size
The result gives you the minimum number of complete responses needed. The calculator also factors in expected response rates to tell you how many people to invite.
For a detailed formula walkthrough, see the Survey Sample Size Guide.
How Many People to Invite?
The calculator gives you two numbers:
- Minimum sample size - the responses you need for statistical validity
- Send survey to - the number of people to invite, adjusted for expected response rate
Typical Survey Response Rates
| Survey Type | Response Rate |
|---|---|
| In-app | 20-40% |
| Email customers | 10-30% |
| Employees | 30-50% |
| Transactional | 15-25% |
| NPS | 10-20% |
Invite formula: People to invite = Required responses ÷ Expected response rate
Example: Need 400 responses, expect 20% → invite 2,000 people.
Common Mistakes
Confusing sample size with population size. Having 10,000 customers doesn’t mean you need 10,000 responses. Around 370 gives you ±5% accuracy at 95% confidence.
Forgetting about subgroups. If you plan to analyze results by region, department, or customer segment, each subgroup needs its own adequate sample size. You might need 370 responses per segment, not 370 total.
Ignoring response rates. Calculating that you need 400 responses but only inviting 400 people guarantees insufficient data. Always divide by your expected response rate.
Over-precision. A ±3% margin requires roughly 2.5× more responses than ±5%. For most business decisions, ±5% is sufficient.
Related Calculators
- NPS Calculator - Calculate and interpret your Net Promoter Score
- CSAT Calculator - Calculate your Customer Satisfaction Score
Learn More
For the full explanation of sample size statistics, formulas, and best practices, see the Survey Sample Size Guide.