We want to decide which icon to show to users: the watchlist star icon or the reading list icon. To inform this decision, we aim to understand how users interact with the watchlist star on both desktop and mobile web.
Usage of watchstar icon for desktop and mobile on enwiki
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | Jdrewniak | T397532 [Hypothesis] FY2025-26 WE 3.3.4 Reading Lists on web | |||
| Open | None | T400355 [EPIC] WE 3.3.4 Reading List Feature design & development | |||
| Resolved | HFan-WMF | T406390 Pull data on watchlist star usage |
Following the instrumentation info that @Jdlrobson-WMF mentioned in T401972#11096518, I didn’t find the event from logged-in users on enwiki in web_ui_actions schema. Instead, I only see menu.watch events from logged-out users on mobile web. Could anyone confirm whether the instrumentation is enabled as expected?
The query
query = ''' SELECT performer.edit_count_bucket, agent.client_platform_family , CAST( performer.is_logged_in AS string) AS is_logged_in , action_source, count(1) AS clicks, COUNT(DISTINCT performer.session_id) AS sessions FROM event.mediawiki_web_ui_actions WHERE year=2025 AND action_source IN ( 'ca-watch', 'ca-unwatch', 'menu.watch', 'menu.unwatch') AND mediawiki.`database`='enwiki' GROUP BY performer.edit_count_bucket, agent.client_platform_family ,performer.is_logged_in, action_source '''
Result
| edit_count_bucket | client_platform_family | is_logged_in | action_source | clicks | sessions |
|---|---|---|---|---|---|
| None | mobile_browser | FALSE | menu.watch | 26421 | 21234 |
Change #1194747 had a related patch set uploaded (by Jdlrobson; author: Jdlrobson):
[mediawiki/core@master] Enable instrumentation of watchstar
Bad news :-(! There is a bug which stops the watchstar being instrumented (above patch will fix it) which blocks this analysis.
Change #1194998 had a related patch set uploaded (by Jdlrobson; author: Jdlrobson):
[mediawiki/extensions/WikimediaEvents@master] Enable instrumentation of watchstar and other links that stopPropagation
Change #1194747 abandoned by Jdlrobson:
[mediawiki/core@master] Enable instrumentation of watchstar
Reason:
https://gerrit.wikimedia.org/r/c/mediawiki/extensions/WikimediaEvents/+/1194998
Change #1194998 merged by jenkins-bot:
[mediawiki/extensions/WikimediaEvents@master] Enable instrumentation of watchstar and other links that stopPropagation
Change #1195049 had a related patch set uploaded (by Jdlrobson; author: Jdlrobson):
[mediawiki/extensions/WikimediaEvents@wmf/1.45.0-wmf.22] Enable instrumentation of watchstar and other links that stopPropagation
Change #1195049 merged by jenkins-bot:
[mediawiki/extensions/WikimediaEvents@wmf/1.45.0-wmf.22] Enable instrumentation of watchstar and other links that stopPropagation
Mentioned in SAL (#wikimedia-operations) [2025-10-09T21:22:56Z] <jdlrobson@deploy2002> Started scap sync-world: Backport for [[gerrit:1195049|Enable instrumentation of watchstar and other links that stopPropagation (T406390)]]
Mentioned in SAL (#wikimedia-operations) [2025-10-09T21:47:54Z] <jdlrobson@deploy2002> jdlrobson: Backport for [[gerrit:1195049|Enable instrumentation of watchstar and other links that stopPropagation (T406390)]] synced to the testservers (see https://wikitech.wikimedia.org/wiki/Mwdebug). Changes can now be verified there.
Mentioned in SAL (#wikimedia-operations) [2025-10-09T22:04:06Z] <jdlrobson@deploy2002> Finished scap sync-world: Backport for [[gerrit:1195049|Enable instrumentation of watchstar and other links that stopPropagation (T406390)]] (duration: 41m 38s)
I have verified the events in the event.mediawiki_web_ui_actions schema. Here’s a summary:
The following analysis is based on data collected from English Wikipedia over a three-week period (October 11–31, 2025). The data represents a 1% sample of English Wikipedia traffic.
1. Number of watch vs unwatch actions
| platform | watch clicks | unwatch clicks | watch : unwatch |
|---|---|---|---|
| desktop browser | 1,291 | 145 | 9:1 |
| mobile browser | 1,068 | 53 | 20:1 |
2. Percentage of unique sessions that clicked watchstar out of all sessions : desktop : 1.41%, mobile: 1.02%
3. Average number of watchstar clicks per session: desktop 7.1, mobile: 5.1
Note: We found five mobile or desktop sessions with over 100 clicks in a day, which might be inflating the average clicks per session.
| client_platform_family | Total sessions | clicks to watch or unwatch | clicked sessions | clicked sessions% | clicks per sessions |
|---|---|---|---|---|---|
| desktop_browser | 14371 | 1436 | 203 | 1.41% | 7.1 |
| mobile_browser | 21539 | 1121 | 220 | 1.02% | 5.1 |
4. Breakdown by edits count bucket : The 1000+ edits bucket shows the highest clicked session rate.
Desktop
| edit_count_bucket | total_sessions | clicks to watch and unwatch | clicked sessions | clicked sessions% |
|---|---|---|---|---|
| 0 edits | 4011 | 230 | 35 | 0.87% |
| 1-4 edits | 1898 | 228 | 12 | 0.63% |
| 5-99 edits | 3880 | 65 | 31 | 0.80% |
| 100-999 edits | 2294 | 53 | 23 | 1.00% |
| 1000+ edits | 2478 | 860 | 102 | 4.12% |
Mobile Web
| edit_count_bucket | total_sessions | clicks to watch and unwatch | clicked sessions | clicked sessions% |
|---|---|---|---|---|
| 0 edits | 4547 | 236 | 84 | 1.85% |
| 1-4 edits | 3185 | 396 | 30 | 0.94% |
| 5-99 edits | 7129 | 351 | 48 | 0.67% |
| 100-999 edits | 3817 | 50 | 21 | 0.55% |
| 1000+ edits | 2941 | 88 | 37 | 1.26% |
5. Breakdown by edits count bucket :The majority of clicks are from article pages.
| client_platform_family | page_type | clicks (watch + unwatch) |
|---|---|---|
| desktop | Article_page | 1271 |
| Article_talk_page | 19 | |
| Others | 146 | |
| mobile web | Article_page | 1031 |
| Article_talk_page | 9 | |
| Others | 81 | |