Hello, hello, hello, and welcome to the second run of the Editor Impact Notifications experiment! You can read more about the first one in the Experimentation Announcement: Impact notifications blog. In short, we experimented with sending a notification to users with an estimation of the impact their edits have in terms of pageviews as an incentive to keep going.
That first experiment sadly ran into some data processing issues, due to which we could not draw a trustworthy conclusion on how this feature did, which prompted us to rerun it.
Experiment details[]
Experiment name: Impact Notifications (still)
Test Launch: Thursday, September 26th
Duration: 4 weeks
Will Be Tested On: 100% of our logged-in users on English wikis using desktop with at least 1 edit
Type: A/B testing, which means that we'll be comparing the editing behavior of the control group A for whom nothing changes with the behavior of group B (and in this case also C) who do receive a notification.
Hypothesis: We still believe that sharing with new editors the impact their first edits are having in terms of pageviews will make editing feel more rewarding for them, feeding into their intrinsic altruistic motivation and causing them to make more edits than they otherwise would.
Other Notes: Just like last time, we will send users a message appearing in their bell notification drawer to alert them of the estimated impact their edits have had on the wiki's pageviews. (Please note that should this experiment end up being productized, ways to show a more accurate number rather than a rough estimate will be investigated.) We will run 2 variants with different text:
First version of the text: The pages you've edited have been viewed # times!
Second version of the text: Your edits have been seen # times! Keep editing, your impact grows with every edit.
The notification is only send once per week and only if the number of pageviews exceeds twenty (20).
Attention: This experiment can conflict with personal dev scripts such as CategorizedNotifications.
Impact of your feedback[]
All the above still pretty much sounds the same as that first experiment, so where did all your feedback make a difference?
While the numbers are still an estimation, the calculation for the data displayed in the notification was adjusted to no longer show "huge" numbers to keep it more realistic.
Edits to non-content spaces (such as your user profile or code pages) will no longer trigger a notification.
And this is the big one, users will be able to opt out of the experiment if so desired. As always we love to hear your actionable feedback and thus encourage you to try it out for a while, but if you no longer wish to receive notifications, you can opt out of the experiment via Special:Preferences, Notifications tab, "Web notifications for all wikis" section.
As always, whether you opt out of the experiment or not, please leave a comment here on the blog or on my message wall to let me know what your overall opinion is on the feature. Only together can we move forward to deliver the best features. Thank you beforehand for your insights and happy editing!
Lostris is a Senior Product Support Manager. When not online, she lives to travel, play roller derby, and escape the world by watching anything that has a good plot, good acting, and preferably a ton of sassy one-liners.
