- Home
- Calculators
- Sample size calculator
Sample size calculator
How many completes you need for a target margin of error, with the formula shown and the B2B caveats explained.
A survey sample size calculator tells you how many completed interviews you need for a result to be precise enough to act on. Enter your confidence level, the margin of error you can live with and the proportion you expect, and it returns the required number of completes. For a single proportion at 95% confidence and a margin of five points, the answer is 385.
Inputs
Required sample size
385
completes at 95% confidence, ±5 points
- Uncorrected n
- 385
- Finite population correction
- not applied
- z score
- 1.960
That is most of your population. When the required sample is close to the whole group, a census is usually cheaper and more defensible than sampling.
Rounded up to the next whole complete. For subgroups, run the calculator on the smallest group you will report separately, not on the total.
How the sample size calculator works
The calculator uses the standard formula for estimating a proportion at a chosen confidence level and margin of error, then applies a finite population correction when you enter a population size.
n = n₀ / (1 + (n₀ − 1) / N)
Where z is the score for your confidence level (1.645 for 90%, 1.960 for 95%, 2.576 for 99%), p is the expected proportion, e is the margin of error as a decimal, and N is the population size. With p at 50%, 95% confidence and a 5 point margin, n₀ is 384.16, which rounds up to 385. That is the number most people recognize.
Why B2B studies usually need a different number
The calculator gives you the n for a single proportion across the whole sample. Most B2B studies report subgroups, for example by company size or industry, and the margin of error for a subgroup depends on the size of that subgroup alone. If you need ±7 points on a segment that is 30% of your sample, run the calculator for the segment, then divide by 0.30 to get the total you need to field.
The second B2B reality is incidence. A required n of 300 at 12% incidence means roughly 2,500 screener starts. That is where cost comes from, and it is why we publish incidence by audience on the audiences page. The incidence rate calculator does that conversion.
Common sample sizes at 95% confidence
| Margin of error | Required n (large population) |
|---|---|
| ±3 points | 1,068 |
| ±4 points | 601 |
| ±5 points | 385 |
| ±7 points | 196 |
| ±10 points | 97 |
Sample size calculator: frequently asked questions
What sample size do I need for a B2B survey?
It depends on the precision you need on the smallest group you will report. For a single overall proportion at 95% confidence and ±5 points, 385 completes. For ±7 points, 196. If you report subgroups, size the smallest subgroup first, then scale the total up by its share of the sample.
Should I use 50% as the expected proportion?
Yes unless you have a prior estimate. Fifty percent produces the largest required n, so it is the conservative choice. If you know the proportion is near 20% or 80%, entering it will reduce the required n.
When does the finite population correction matter?
When your sample is more than about 5% of the population. For a C-suite study where the reachable population is 640 people, the correction reduces the required n noticeably. For a general small business study, it makes no practical difference.
Does a bigger sample fix bad sample?
No. A larger n narrows the margin of error around whatever the sample measures. If a fifth of the respondents do not hold the role they claim, the estimate is biased and more of it makes it more confidently wrong. Verification comes before size.
Send the spec. Get real numbers back.
Audience, market, target n, expected interview length. Feasibility the same business day for standard audiences, with reachable counts, an estimated qualification rate we believe and a quoted cost per complete that stays fixed for the agreed brief.