Page MenuHomePhabricator

Randomise content for users on Microsite
Closed, InvalidPublic

Description

Background

Wiki-highlights experiment is an A/B test, where half the users(treatment group) will consume wiki-highlights and the other half(control group) will consume long-text articles. This content needs to be randomised ensure that we eliminate bias & there's statistical validity of the test.

*Category-level randomisation:*

  1. Each user will see a different long-text article category on the category page.
  2. If a user returns to the category page, they should see the same category as before during the session.
  3. If a user closes the session and comes back, they will see a different category at the top.
Design
Acceptance Criteria
  • Each time a user on either treatment or control groups lands on the category page, the same category is displayed at the top during an active session.
  • When a user ends a session & returns to the category page, they should see a different category displayed at the top.
Test Scenarios
Open questions

Event Timeline

PWaigi-WMF renamed this task from Randomise content for users for wiki-highlights experiment to Randomise content for users on Microsite.Aug 10 2023, 6:01 PM
PWaigi-WMF triaged this task as High priority.
PWaigi-WMF updated the task description. (Show Details)

Dropping this ticket as the work has been captured under ticket T345014