Beginning in Q4 and carrying over into Q1, we will be conducting an A/B test using the new experimentation platform, as a means of determining the optimal donation entry points on both mobile and desktop, as well as helping to test the experimentation platform as the first "real" internal users
Requirements
- an A/B test is run on a TBD population of anonymous users
- on both mobile and desktop
- with the primary metric measured being actual donations (quantity) and auxiliary metric being clickthrough rate of the entry point
- the data of which is then analyzed to determine the "winning" groups on both platforms
- after which the experimental code is removed to make these changes permanent
Design
I'm sure there are more up to date designs somewhere, but the most recent versions I could find are here: T388149
Measurement plan
https://docs.google.com/document/d/1zIVct6RC8hoSCX5yeqsvSI_aFSRARj1JGgwX3pkRC2Q/edit?usp=sharing
Instrumentation spec
https://docs.google.com/spreadsheets/d/1PLbiTF160-jkqqYMN9HBVyXsf5i0FatKyXsFZNO-cSc/edit?usp=sharing