Data sampling is a standard statistics technique used to select, process, and analyze a representative subset of a population. It is also used to identify patterns and extrapolate trends. Sampling is used, for instance, for political or opinion polls. If a researcher wants to determine the most popular way of commuting to work in the US, they won’t need to talk to every American citizen. Instead, they can select a representative group of 1,000 people, hoping it will be enough to make the results accurate.
In web analytics, sampling works similarly. Only a subset of your traffic is selected and analyzed, and that sample is used to estimate the overall results.
Sampling in analytics has its advantages and applies in certain situations. However, using it automatically without knowing the consequences of working on a sample may cause problems. These include report inaccuracy.
What is data sampling and how does it work?
Raw data and sampled data: How to ensure accurate data
Compare 7 free web analytics platforms (product analytics included)