A/B testsMeasure lift, with the confidence to call it
Read the variant, the lift and the confidence together, and be told plainly when a result is not callable yet.
Askacme.com LiveFor E-commerce
This runs on your own data in under a minute.
What it is good for
The failure mode of testing is calling a winner early. This reports lift beside confidence, and when the sample is too small it says so instead of picking a side. Because the assignment is session-stable and person-resolved, a visitor who returns is still in the variant they started in.
- Knowing when a result is safe to ship
- Measuring lift on a real conversion, not a proxy click
- Keeping people in their variant across sessions and devices
- Holding the line when a test is flat
How different businesses use a/b tests
The report is the same. What changes is the question you point it at.
E-commerce
Test the checkout on the order, never on the click.
Control against variant, conversion event Completed order, 95% threshold.
You ship changes that moved revenue rather than changes that moved a micro-metric.
Give it your URL. It does the rest.
Free to start, no card. Tracking, attribution and your first dashboard in under a minute.
Want a hand? Get a live demo