We recommend testing at least 160 respondents per survey for statistical reasons, however the minimum number required to complete the test is just 10 respondents. The number of respondents depends on various factors such as your target group, the total population of your target group and the incidence rate. We also advise oversampling by at least 10%. This is due to the fact that some respondents will be excluded in the cleaning process (e.g. due to too fast or too slow responses).
Updated on 16 October 2019