PaneliaTools

Sample size for 10,000 people

10,000 people — a mid-size CRM base, a large company's workforce, a town: at this scale the finite population correction still helps, but modestly. You need 370 respondents for ±5% at 95% confidence, versus 385 for an infinite population. The saving is real (15 respondents) but we're entering the zone where population size stops being a lever.

That's this page's practical lesson: between 10,000 and infinity, the required sample varies by less than 4%. If your base is 'around 10,000', don't bother measuring it precisely — the error is negligible. Invest instead in recruitment representativeness, which weighs far more on final quality.

Confidence level

95% is the market research standard. Z-scores: 1.645 · 1.96 · 2.576 (NIST statistical tables).

The acceptable gap between your sample and reality. ±5% is the most common choice.

If unsure, leave 50%: it's the worst case, requiring the largest sample.

The total number of people in your target. Above ~100,000 the impact is negligible: leave empty.

Respondents needed

370

You need 370 respondents for a 95% confidence level with a ±5% margin of error.

Export:

How many respondents per precision level?

Precision is expensive: going from ±5% to ±2% multiplies the sample by 6.

101001,00010,0001%3%5%8%10%Margin of error

Summary table

Sample size for the most common combinations.

Summary table
Confidence± 3%± 5%± 10%
90%70026468
95%96537096
99%1,557623164

Sample size: done. Now, the fieldwork…

Traditional fieldwork takes 6 weeks and $10,000. Panelia simulates 300+ calibrated respondents in 10 minutes.

Simulate my study

Frequently asked questions

How many respondents for 10,000 people at ±3%?
965 respondents at 95% confidence (versus 1,068 for an infinite population). The FPC saves about 10% of the sample at this precision level.
From which population size does the FPC become negligible?
Beyond ~20 times the required sample, the gain drops below 5%. For a standard survey (n₀ = 385), population stops mattering beyond ~8,000–10,000 people.
Should I stratify my 370-person sample?
If your population has very different segments (jobs, regions), a proportional stratified draw improves representativeness without changing the total size. Each stratum analyzed separately must however be sized on its own.
10,000 customers but only 2,000 valid emails: which N?
The relevant N is the population you CAN reach: 2,000. Watch for coverage bias — customers with an email may differ from the rest; flag that limit in your conclusions.