Conjoint.app, an educational tool by Subconscious.ai
Causal experiments for business decisions
Change one thing. Compare two similar groups. Measure what follows.
Correlation can hide another cause
Ice cream sales and shark attacks can rise together because hot weather raises both. A hidden factor can make two numbers move together even when one did not cause the other.
Random assignment creates a fair comparison
Assign people to two groups by chance. Keep everything the same except the change being tested. The difference between the groups estimates the causal effect of that change.
A confidence interval shows uncertainty
The estimate is one number. The confidence interval shows a range of results that fit the evidence. Larger tests usually produce a narrower range.
Subconscious applies the method to buyer choices
Subconscious runs controlled discrete-choice experiments on synthetic populations and reports causal effects with confidence intervals.