In T253902: Variant tests: revisit edit completion for A vs. B, we decided that Variant A is the stronger variant of A vs. B, because it causes more newcomers on mobile to edit (while leaving the same number editing on desktop).
Until we deploy T246533: Variant tests: "initiation part 2" test (C vs. D), we should stop our Variant A vs. B experiment, and instead start giving Variant A to all newcomers.
| Subject | Repo | Branch | Lines +/- | |
|---|---|---|---|---|
| GrowthExperiments: End A/B test | operations/mediawiki-config | master | +0 -9 |
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | MMiller_WMF | T238887 [EPIC] Growth: Homepage variant tests | |||
| Resolved | • nettrom_WMF | T238888 Variant tests: "initiation" test (A vs. B) | |||
| Resolved | Catrope | T254413 Variant tests: switch all newcomers to Variant A |
Change 603548 had a related patch set uploaded (by Catrope; owner: Catrope):
[operations/mediawiki-config@master] GrowthExperiments: End A/B test
Change 603548 merged by jenkins-bot:
[operations/mediawiki-config@master] GrowthExperiments: End A/B test
Mentioned in SAL (#wikimedia-operations) [2020-06-08T18:28:19Z] <catrope@deploy1001> Synchronized wmf-config/InitialiseSettings.php: End GrowthExperiments homepage A/B test (T254413) (duration: 00m 57s)
Checked on desktop and mobile on betalabs - no issues are found - now all new users will get variation A