“Which screenshot makes you want to try the product?”
95% interval 33%–79%
95% interval 21%–67%
The leader's interval still crosses 50%, so this sample can't separate the two.
Reason tags are optional, so these add up to less than the total vote count.
Reading the comments…
Showing the nav tells me the app has depth. B could be a single screen.
Two versions, one question, and people who have no reason to be polite about it.
A is ahead at 57% of 14 boops, but the 95% interval (33%–79%) still includes 50%. This is a lean, not a result.
Themes
More useful
Tagged 4 times — 29% of everyone who answered.
Cleaner
Tagged 2 times — 14% of everyone who answered.
Easier to understand
Tagged 2 times — 14% of everyone who answered.
Built from the counts alone. Add an OPENAI_API_KEY to have written comments grouped into themes.