How Many Survey Responses Do You Need for Reliable Results?
Short answer, for a general audience and a typical 95% confidence level with a 5% margin of error: about 385 responses, regardless of how large the total population is, and fewer than that if the population itself is small. The table below works as a quick sample-size lookup; the sections after it explain where the numbers come from and when you can reasonably use far fewer.
What determines survey sample size?
Three things set the number of responses you need: population size, confidence level, and margin of error:
- Population size: the total number of people you could possibly survey (all customers, all employees). Counterintuitively, the required sample barely grows past a certain population size (see the table below).
- Confidence level: how sure you want to be that your sample reflects the full population. 95% is the standard default.
- Margin of error: how much the real answer could differ from what your sample shows. 5% is the standard default; tighter margins (e.g. 3%) need meaningfully more responses.
How many responses do you need, by population size?
This table is the survey sample size calculation most people are looking for. Read down to your population size to find the responses needed at 95% confidence and a ±5% margin of error, or use the sample size calculator to put in your own numbers:
| Population size | Responses needed |
|---|---|
| 100 | 80 |
| 500 | 218 |
| 1,000 | 278 |
| 5,000 | 357 |
| 10,000 | 370 |
| 100,000+ | 385 |
Based on the standard finite-population-correction formula (Cochran / Krejcie & Morgan), assuming maximum variability (p = 0.5), the most conservative and most commonly cited version of this table.
When can you use a smaller sample?
These numbers are for statistical inference: being able to say "72% of customers agree, ±5%" with real confidence. A lot of survey work doesn't need that:
- Exploratory or qualitative feedback. If you're reading open-text answers for themes, not computing a percentage, 20–30 responses is often enough to see patterns repeat. See how to analyze open-ended responses.
- Directional signal, not precision. Deciding between two product directions rarely needs statistical rigor. A clear lean in 40 responses is usually enough to act on.
- Tracking change over time. A consistent, smaller sample measured the same way every quarter can show a trend clearly, even if any single wave wouldn't stand alone statistically. This is how metrics like NPS and CSAT are usually tracked.
What if you can't reach the target sample size?
Report what you have honestly rather than not at all. State the actual response count next to any percentage, note that the margin of error is wider than the standard table implies, and lean on trends across multiple survey waves rather than betting everything on one wave's precision. A smaller, honestly-labeled sample is more useful than an unlabeled one that quietly overclaims certainty.
If the shortfall is people not answering, the fixes in how to raise your survey response rate are the place to start. And remember that sample size only governs precision, not bias. A large sample built on leading or double-barreled questions is confidently wrong, not reliable.
Frequently asked questions
How many survey responses is statistically significant?
For a large population at the standard 95% confidence level and a 5% margin of error, about 385 responses. That number barely changes once the population is large, so 385 is a reasonable target for most public or large-audience surveys.
What is a good margin of error for a survey?
5% is the standard default and is fine for most decisions. Tightening it to 3% roughly doubles the responses required, so only pay for that extra precision when a decision genuinely turns on it.
Do you need a bigger sample for a bigger population?
Only up to a point. The required sample grows quickly for small populations, then plateaus. A population of 10,000 and one of 10 million need almost the same sample (around 370–385) at 95% confidence and a 5% margin of error.
How many responses do you need for qualitative feedback?
Far fewer. If you're reading open-text answers for recurring themes rather than computing a percentage, 20–30 responses is often enough to see patterns repeat.
- NPS vs CSAT: Which Survey Metric Should You Use? · What each score actually measures, how they're calculated, and when to run one, the other, or both.
- How to Write Survey Questions That Get Honest Answers · Seven concrete ways question wording quietly biases responses, and how to fix each one.
- How to Analyze Open-Ended Survey Responses · A practical method for turning free-text answers into themes you can count, quote, and act on.
- What Is a Good Survey Response Rate, and How Do You Raise It? · How to calculate response rate, why benchmarks mislead, and the changes that actually get more people to answer.
- Likert Scales: How to Design and Analyze Them · 5 or 7 points, whether to include a neutral midpoint, and how to report the results without averaging away the story.
- Likert Scale Examples: Answer Options for Every Kind of Question · Ready-to-copy 5- and 7-point labels for agreement, likelihood, satisfaction, frequency, importance and more, and which question each one fits.
- How to Present Survey Results to Leadership · How to structure a survey report so the people reading it can make a decision, not just look at charts.
- How to Check an AI Summary of Your Survey Results · Eight checks that catch the mistakes AI tools make when they summarize survey data: missing bases, mislabeled groups, invented rankings and more.
- Survey Report Example and Template: What a Good Report Contains · A real survey report walked through section by section, plus a fill-in template you can copy for customer, employee and market research surveys.
- Sample size calculator · Put in your own population, confidence and margin
SurveyService's free plan shows your first 50 responses in full results and AI summaries; paid plans raise that to 1,000–2,000. See plans.