Page MenuHomePhabricator

Homepage: analysis of visits to user and user talk pages
Closed, ResolvedPublic

Description

This task is a specific analytics request from the EditorJourney data that will help us predict how often users will make it to their homepage without additional encouragement. It is an offshoot of the EditorJourney initial report, which gave numbers on the percent of newcomers who visit their User or User Talk page in their first 24 hours (34% in Czech, 39% in Korean). This task asks for more specific drilldowns.

  • What percent of users visit their User page in their first 24 hours?
  • What percent of users visit their User Talk page in their first 24 hours?
  • What percent of users visit both in their first 24 hours?
  • From the above numbers, we'll also know the percent who visited just one of the pages or neither of the pages.

These numbers should be split by platform (mobile/desktop) and wiki (Korean, Czech, Vietnamese). They should exclude any subpages.

Event Timeline

Restricted Application added a subscriber: revi. · View Herald TranscriptMar 5 2019, 5:50 AM
MMiller_WMF renamed this task from Homepage: visits to user and user talk pages to Homepage: analysis of visits to user and user talk pages.Mar 6 2019, 9:46 PM

Using all available data from EditorJourney, excluding known test users, and allowing "visit" to be either viewing the page or opening the editor, the results are as follows:

Overall:

WikiVisited UserVisited User talkVisited bothVisited neither
Czech29.3%14.0%7.9%64.6%
Korean15.8%29.3%10.8%65.7%
Vietnamese20.4%13.6%7.5%73.5%

Split by whether the account was registered on desktop or mobile:

WikiDesktop/mobileVisited UserVisited User talkVisited bothVisited neither
CzechDesktop30.5%17.0%9.8%62.3%
CzechMobile25.8%5.9%2.9%71.2%
KoreanDesktop15.2%30.0%11.1%65.8%
KoreanMobile17.8%26.8%9.8%65.2%
VietnameseDesktop22.3%16.9%8.9%69.7%
VietnameseMobile17.1%7.7%5.0%80.2%

Reassigning to @MMiller_WMF for review.

Thanks, @nettrom_WMF -- this helps. So depending on the wiki, we can expect between 15% and 30% of newcomers to encounter the homepage on their first day (with no additional features that drive them there). My main takeaway is that we should prioritize feature work that drives users there, so that more than that smallish percentage get exposed.