Sample Size Calculator

Choose your confidence level and desired margin of error to see how many respondents you need.

Population size (optional)

The total number of people you could possibly survey. Leave blank if unknown or very large.

Confidence level

Margin of error

Respondents needed

Assumes a 50% response distribution (the most conservative estimate). Formula: n = z²p(1-p)/e², adjusted for population size when provided.

Create your survey

What is sample size?

Sample size is the number of people who need to respond to your survey for the results to reliably represent your entire target audience — your "population." Survey too few people and your results could easily be skewed by chance; survey far more than necessary and you're spending extra time and budget for marginal gains.

Data you need to calculate sample size

Population size — the total number of people you could possibly survey (e.g. all your customers, or all employees).
Margin of error — how much error you're willing to tolerate, expressed as a plus-or-minus percentage. See our Margin of Error Calculator.
Confidence level — how confident you want to be that your results reflect the true population (90%, 95%, and 99% are the most common).
Response distribution — an estimate of how responses will split. 50% is the safest default since it produces the largest, most conservative sample size estimate.

How to calculate sample size

n = z² × p(1-p) / e²

z = z-score for your confidence level · p = response distribution (0.5 default) · e = margin of error (as a decimal)

If you know your total population size, that raw number is then adjusted downward with a finite population correction — which the calculator above applies automatically once you enter a population size.

Confidence level	Z-score
90%	1.645
95%	1.96
99%	2.576

Worked example 1

A large, unknown population, 95% confidence, ±5% margin of error:

n = 1.96² × 0.5 × 0.5 / 0.05² = 384.16
You need about 385 respondents.

Worked example 2

A 300-person company survey, 90% confidence, ±10% margin of error (finite population applied):

Raw n₀ = 1.645² × 0.5 × 0.5 / 0.10² ≈ 68
Adjusted for a population of 300: n ≈ 56
You need about 56 respondents out of 300 employees.

What's a good sample size for your survey type?

Statistical rigor matters more for some surveys than others. Here's rough guidance by use case:

Survey type	Guidance
Customer satisfaction / NPS	Aim for statistical significance (use the calculator) if you're tracking trends over time or comparing segments.
Employee engagement	With a small, known population, survey everyone if possible — response rate matters more than sample-size math.
Market research	Statistical rigor is important; under-sampling risks basing real business decisions on noise.
Academic / scientific research	Follow your field's accepted methodology and confidence standards closely — usually 95% or higher.
Quick internal polls	Precision is less critical; a smaller, directional sample is often good enough.

Common sample size mistakes to avoid

Confusing sample size with response count. If you need 385 responses and expect a 20% response rate, you'll need to invite roughly 1,925 people.
Ignoring subgroup analysis. If you plan to break results down by segment (region, age group, plan tier), each segment needs enough responses on its own — not just the overall total.
Assuming a bigger sample always means better data. A large but biased or unrepresentative sample can be less reliable than a smaller, well-targeted one.
Not accounting for drop-off on long surveys. Longer, open-ended-heavy surveys see higher abandonment — pad your invite list accordingly.

Frequently asked questions

What sample size is considered "large"?

There's no universal cutoff, but as a rough guide, samples of 30+ are generally treated as large enough for standard statistical methods to apply reliably. For population-level surveys, 384+ is the common benchmark at 95% confidence and ±5% margin of error.

Do I need a different sample size for each question in my survey?

Technically yes — each question's effective sample size is the number of people who actually answered it, which can shrink for optional or later questions due to drop-off.

What if I don't know my exact population size?

Leave the population field blank. The calculator will treat your population as effectively unlimited, which produces a slightly more conservative (larger) sample size recommendation — a safe default.

How does margin of error relate to sample size?

They're inversely related — a smaller desired margin of error requires a larger sample size, and the relationship isn't linear (halving your margin of error roughly quadruples the sample size needed). Use our Margin of Error Calculator to explore the trade-off.