Statistical power
The probability that a test detects a real effect of a given size when one exists.
Tests are commonly planned at 80% power: if the variant really is better by at least the minimum detectable effect, the test will show it eight times out of ten.
Underpowered tests produce false negatives. They call a good idea a loser simply because too few visitors took part. More power needs a larger sample size.
Put it to the test on your store
Test prices, themes, pages, shipping and checkout on Shopify, then ship the winning variant.
