EN
FR
ES
DE
A confidence interval expresses the margin of uncertainty around a measurement. The Wilson method is suited to proportions calculated on small samples, such as a brand's presence rate measured across a limited number of AI answers.
Measuring uncertainty, not just a number
AI answers vary from one run to the next. A mention rate measured across a few dozen answers is therefore not an absolute truth. The Wilson confidence interval frames that rate with a range: "the brand appears in 40% of answers, give or take 8%".
Why Wilson rather than a classic calculation?
The Wilson method stays reliable on small samples and near the extremes (0% or 100%), where the classic formula produces absurd results. That matters when measuring the visibility of a brand that is still rarely cited.
In the AGS
The AGS displays this interval on presence scores: uncertainty is shown, not hidden. It is one of the pillars of transparency in our methodology, alongside inter-rater reliability.
Every question asked to ChatGPT without your name in the answer is a competitor recommended instead of you — measured across 6,820 real AI answers.