Skip to main content

Optimization and personalization

Improve checkout performance with A/B testing (experiments), profit-optimized delivery and checkout translations. Includes how test results, variants and statistical significance work.

1 author23 articles
A/B testing
What does Ingrid A/B test do?
What Ingrid features can we run A/B testing for?
What dependent variables/outcomes can be analyzed through the A/B test?
What does the "Original" variant represent?
I want to change the settings for the "Original" variant - how do I do that?
What happens if I change the main site configurations when an A/B test is run?
What happens if I do changes in Region A, while having an active AB test on region B?
Which A/B test variant a shopper sees
Can I set criteria or a date when my experiment should end?
How frequently are the A/B test results updated?
How many sessions should the test receive before ending it?
What is statistical significance, and why does it matter?
Why aren't any of my variants significant?
Why does the significance level change between being significant and non-significant?
Will running a test longer make the significance go above 95%?
The significance level is at 92%, is it close to being significant?
What are common mistakes to avoid in A/B testing checkouts?
Is it possible to interact with AB testing through any public APIs?
FAQ Experiments
Profit-optimized delivery
Setting up translations
The "Original" have a box next to it saying "out of sync" - what does it mean?