Report what changed in AI answers
The same two rates, 40% and 60%, support different claims depending on which questions you asked and what you compared them with. Use these fictional examples to choose wording for your report.
Choose your comparison
What you measured
- Original questions
- 8 of 20
- Different questions
- 12 of 20
Each count is the number of answers that named the brand or linked to its official site.
What you can report
The brand appeared in 8 of 20 answers to the original questions and 12 of 20 answers to a different set of questions.
Rerun the original questions under the same conditions to measure change for that set.
What you measured
- First round
- 8 of 20
- Second round
- 12 of 20
- Gains and losses
- 6 gains, 2 losses
Both rounds used the same questions and measurement settings.
What you can report
Across 20 matched questions, brand presence rose from 40% to 60%. Six answers gained presence and two lost it.
Say when you measured and which answers were matched. Check other explanations before crediting an update for the gain.
What you measured
- Monitored set
- 8 of 20 to 12 of 20
- Comparison set
- 7 of 20 to 9 of 20
- Gap in changes
- 10 percentage points
Each set used the same questions before and after, with settings held steady.
What you can report
Presence rose 20 percentage points in the monitored set and 10 in the comparison set. The gap between those changes was 10 points.
Compare the groups’ prior trends and check for other differences before interpreting the gap as an effect.
Keep the counts with the claim
Save the question set, date range, engine settings, market, and exact rule for presence. The VisibilityOS study reported 82 presence changes across 817 matched comparisons and documented the matching rules alongside that result.