Noise is a privacy control, designed to prevent accidental or deliberate identification of individuals through the results of insight queries.
Noise defends against a category of attacks built around changing the input data incrementally, and observing change in the results.
Without noise, a malicious user could gradually add rows to a dataset until they see a statistic move over the rounding threshold - which would reveal the original value of the statistic.
To apply noise, InfoSum Platform modifies every statistic by a small but unpredictable amount. Repeating exactly the same input will result in the same noise (and therefore the same result). But any change to the input will result in a different amount of noise, and will therefore vary the result in an unpredictable way.
By default, ±1-2% noise is injected into an activation file so the final number can diverge between ±1-2% from the actual value. For this reason, privacy safe results can be up to ±0.01-2% different to the actual result with noise.