What Is a Good Sample Size for a Survey?
There is no single number that is always a good sample size. The right target depends on how precise you need the result to be and how the survey will be used.
Why there is no universal answer
A good sample size for one study can be excessive for another. A lightweight feedback pulse and a major market research project do not need the same level of certainty or precision.
That is why sample size should match the stakes of the decision.
A common benchmark
For large populations, 95% confidence and a 5% margin of error often produce a target somewhere around the mid-300s. That is why many survey planners see numbers near 385 so often.
But this is only a common starting point, not a rule that fits every survey.
What makes a sample size good
A good sample size is large enough for reliable conclusions, realistic for the project, and appropriate for subgroup analysis if you need comparisons.
If the survey result will be sliced by region, department, customer tier, or device type, those subgroups may need their own planning.
How to judge whether a target is truly good
A good sample size is one that supports the decision you need to make with enough precision to be credible. That usually means the number is neither too small to trust nor so large that it wastes time and budget.
The label good becomes weaker when people ignore audience structure. A sample that is good for the full population can still be poor for comparing teams, regions, or customer groups.
- Match the target to the stakes of the decision
- Check whether subgroup analysis changes the requirement
- Balance precision against cost and fieldwork reality
- Avoid treating common benchmarks as universal rules