VisibilityOS by PodPitch AI answer research

AI search control group calculator

Compare how often AI answers mentioned your brand before and after a change. Use an unchanged set of prompts to check whether mention rates also shifted elsewhere.

Enter the number of answers that mentioned your brand and the total number of completed answers. The starting counts are fictional.

Test prompts

Questions your change was meant to affect.

MentionsAnswers Before After

Comparison prompts

Similar questions outside the scope of your change.

MentionsAnswers Before After

Gap between the changes

+12.5

percentage points

Test prompts
37.5% to 62.5% +25.0 points
Comparison prompts
25.0% to 37.5% +12.5 points

Build a fair comparison

Choose both sets before the change. Keep the exact wording, model, search mode and market fixed across rounds. Run the two sets on the same dates. Look at several earlier rounds to check whether their rates usually move together.

Read the gap

Subtract the comparison group’s change from the test group’s change. The result is descriptive. If the sets already had different trends or the change affected both, the gap can mix those effects. World Bank explanation of the method.