In the 2024-2025 fiscal year, the Editing, Growth, iOS, and Research Teams will work together to increase constructive activation. [i]
In T371726, we are working to set the percentage by which we'll strive to improve the baseline constructive activation rate (T360829).
In this ticket, the Editing, Growth, iOS and if applicable, Research Teams will look back at the impact past interventions [ii] they've deployed have had on improving constructive activation to inform the improvement target we'll ultimately set in T371726.
Past interventions
The list below includes the interventions that have sought to impact constructive activation (or related/analogous metrics) and effect they were proven to cause. Please see the Google Sheet version for more detailed documentation.
| Intervention name | Description | Analysis date | Platform(s) | Target Audience | Observed impact[1] | Notes/link(s) | |
|---|---|---|---|---|---|---|---|
| Newcomer tasks | In November 2019, the Growth team added the "newcomer tasks" feature to the newcomer homepage, which provides a feed of suggested articles to edit tailored to the area of interest of the newcomer. | November 2019 through May 2020 | Mobile web and desktop | Reviewed 97,755 new accounts logged on four pilot wikis: Arabic, Vietnamese, Czech, and Korean Wikipedias. | 26.7% increase (16.1% → 20.4%; 4.3pp) | Newcomer Tasks/Experiment Analysis | |
| "Add a link" structured task | In May 2021, the Growth team added the "Add a Link" structured task to the "newcomer tasks" feature on the newcomer homepage. | May 2021 through October 2021 | Mobile web and desktop | Reviewed 130,179 new accounts data logged on 10 pilot wikis: Arabic, Bengali, Czech, Vietnamese, Russian, French, Polish, Romanian, Persian, and Hungarian | 16.9% increase (20.7% → 24.2%; 3.5 pp). Note this is desktop and mobile web combined. Per platform splits shown in this image where it appears that mobile web rates increased from around 15% in control to 21% in treatment | Add a link/Experiment analysis | |
| "Add an image" structured task | In late November 2021, the Growth team added the "Add an Image" structured task to the "Suggested Edits" module on the Newcomer Homepage | July 2022 through October 2022 | Mobile web and desktop | Reviewed 148,412 new accounts looged on seven pilot wikis: Arabic, Bangla, Czech, French, Persian, Portuguese, and Turkish. | Desktop: 1.9% increase (21.6% → 22.0%; 0.4pp); Mobile: 17.0% increase (16.6% → 19.4%; 2.8pp) | Add an image/Experiment analysis | |
| Impact Module | Changes were made to the Impact module to provide new editors more context about their impact, as well as encourage them to continue contributing | February 2023 | Mobile web and desktop | Four pilot wikis: Arabic Wikipedia, Bengali Wikipedia, Czech Wikipedia, and Spanish Wikipedia | 22.5% activation rate for both the new impact and old impact following improvements and fixes deployed. This was the expected result as the old empty state design is very similar to the new empty state design, so we would expect to see no change in the Activation rate. (investigation conducted because there was a initially an observed decrease on mobile web from (17.1% to 15.2%) | Impact module/experiment results | |
| Newcomer experience marketing experiment | As part of the Wikimedia Foundation's FY21–22 pilot project, Communications and Growth teams tried various tactics to acquire and retain new Wikimedia editors on the Spanish Wikipedia. This included a new on-wiki sign-up landing page and welcome emails. | May 2022 through July 2022 | Desktop and mobile web | Market: Spanish-speaking South, Central, and North America. Target audience: 18–45, at least college educated | New sign-up page : 1 contructive activation (control) -> 10 constructive activations (test) (Numbers too low to establish statistical significance); Welcome emails: No effect; 0.5% increase (19.8% → 19.9%; 0.1 pp) | Newcomer Experience Marketing Report | |
| Thank you pages and thank you banners | Project undertaken as an experiment, as a collaboration by the Growth and Fundraising teams at the Wikimedia Foundation. During this project, we leveraged existing fundraising activities to ask readers and donors to create accounts. | May 2022 through July 2022 | Thank you page (desktop); Thank you banner (mobile web) | Targeted donors in Latin America, India, and South Africa; Invitation to edit sent via emails and banners on enwiki, eswiki, and ptwiki | Thank you page: Only 4.6% of accounts created started editing within 24 hours of creating their account. Per region activation rates range from 2.1% in Latin America (enwiki) to 8.2% in Brazil (ptwiki); Thank you banner: 16.8% of accounts were constructively activated. Per region activation rates range from 7.3% Latin America (enwiki) to 25.3% in Brazil (ptwiki) ; Organic activation rate was estimated to be between 26-36% on these wikis. | Newcomer Experience Pilot Project/Thank you emails and banners | |
| Newcomer Homepage with Newcomers Taks module | Experiment evaluated the impact of providing a central place where newcomers can learn how peer production works and find opportunities to contribute, as a solution for attracting and retaining newcomers | February 2021 - May 2021 | Desktop and mobile web | 244,060 accounts registered on 27 non-English Wikipedia wikis | Overall: 1% increase in the odds of activation but depends on newcomer context. Some insights: Users that were editing at time of registration saw a significant decrease in constructive article activation if they were also in the treatment group. For ssers who said they registered to read wikipedia, there was a significant increase in activation for respondents who said they created their account to read Wikipedia. | Increasing Participation in Peer Production Communities with the Newcomer Homepage | |
| 2019 Growth Help Panel | The help panel was first deployed to newcomers in January 2019. The goal of the Help Panel is to provide users with easier ways to get help without them having to leave the editing context. Activation rate (not constructive activation) was identified as a target metric for the project. | January 2019 - February 2019 | Desktop and mobile web | 3,848 accounts in Czech Wikipedia, 3,690 accounts in Korean Wikipedia, and 5,283 accounts in Vietnamese Wikipedia. | As part of the analysis, the team used historical data to calculate an estimate of what the average activation rate for each wiki has been and the effect size we would be able to detect based on the number of users in our dataset. Results: | https://www.mediawiki.org/wiki/Growth/Analytics_updates/Help_Panel_editor_activation_rate | |
| Image Recommendations | The Android Team added Image Recommendations as a suggested edit type to the Android app | October - November 2023 | Android | #TODO | 4.8% m-o-m increase in the number of people using Suggested Edits each day | Android Add Image Recommendations 30 Day Analysis | |
| Image Recommendations | The iOS Team added Image Recommendations as a suggested edit type to the iOS app | May 2024 | iOS | TODO | 8.0% increase in the number of unreverted edits made to the main namespace | iOS Image Recommendations 30 Day Analysis | |
| Native full-page editor | The iOS Team improved the full-page editing interface | iOS | TODO | 18.8% increase in the average number of edits published/editor (6.0 → 7.1 edits/editor) | |||
i. Where "constructive activation" refers to the percentage of brand new editors who publish ≥1 edit in the main namespace on a mobile device that is not reverted within 48 hours, as measured by controlled experiments.
ii. Where "past interventions" is relatively open-ended and includes all interventions that sought to cause a change in a metric that is similar enough to how we've defined "constructive activation" in the context of the WE 1.2 KR: "Newcomers making their first edit to an article in the main namespace of a Wikipedia projects on a mobile device (mobile web, Android, or iOS app) within 24 hours of registration, and that edit not being reverted within 48 hours of being published.