Product and Technology Advisory Council/February 2025 draft PTAC recommendation for feedback/Mobile experiences

Why, and why now?

People using mobile devices account for around 60% of Wikipedia traffic, and that number is only growing as desktop traffic declines. In the Global Majority, around 70% of people exclusively access the Internet using mobile devices. However, based on data from recent months, only around 12% of successful edits are accomplished on a mobile device. Because Wikimedia's mobile infrastructure is not up to the task, we are excluding large swaths of the global population from participating in the sum of all human knowledge. Right now, Wikipedia is not the encyclopedia anyone can edit; it is the encyclopedia anyone with the privilege of access to a desktop device can edit.

Mobile is a particularly urgent priority because new audiences who are coming online are primarily mobile. People are also increasingly moving to mobile devices (as desktop device usage decreases). The segment of people who are facing challenges in participating because they primarily or exclusively use mobile devices is only increasing, and so we need to address those challenges now.

Problem statement

Despite mobile accounting for around 60% of Wikipedia's traffic, it contributes only around 12% of successful edits. Challenges in mobile editing make it difficult, if not impossible, for some mobile editors to contribute, particularly those who do not have any access to a desktop device that they can use to complete edits they began on mobile devices. We are dramatically limiting the number of people who can fully participate in the Wikimedia movement by effectively excluding those without access to desktop devices.

We have identified several areas of possible focus:

Improving/fixing mobile web editing user experiences (e.g. Visual Editor)
Improving the experience for new editors using mobile devices
Improving the experience for editors from the Global Majority, who disproportionately use mobile devices

What the metrics and data are telling us

Analyzing data from recent months, showed mobile web pageviews accounted for around 60% of Wikipedia's traffic, but only around 12% of successful edits are made with mobile devices. In the Global Majority, 70% of people exclusively use mobile devices to access the Internet.

Mobile readership globally has increased and continues to trend upwards. Mobile web pageviews overtook desktop pageviews in around 2017:

(Note: Dates formatted as MM/DD/YYYY)

Relatively steady pageviews across all devices during that same period suggest that this reflects a shift from desktop to mobile devices.

(Note: Dates formatted as MM/DD/YYYY)

There is also a dramatic discrepancy between desktop and mobile web when examining how many editors successfully navigate the process from beginning an edit to successfully saving a constructive (not reverted) edit. Only 6% of editors who begin to edit on mobile web even attempt to save their edit, compared to 24% on desktop. Only 4% of mobile web editors successfully save a constructive (non-reverted) edit, compared to 19% of desktop users.

Source: WMF Movement Insights Team Calculations.
Note: 'Initiates editor' (inits) is the baseline at 100%, and each percentage represents the proportion of initiations that progress to the corresponding stage of the funnel; Data represents a sample of edit actions from October 1st to 15th 2024; Includes data from all wikis; Excludes edits made by bots; The time delay to reversion is ~60 days; Data for mobile app edits are not included. The inits are an approx 1% random sample of data from October 1st to 15th 2024: desktop inits: 67,803 and mobile web inits: 47,975.

What would success look like?

People can successfully contribute on mobile web without hitting technical barriers throughout the funnel. This can be measured by looking at the discrepancy between desktop and mobile devices in the user editing funnel at every step along the way. Potential metrics:
- Increase in % of successfully saved edits made on mobile web
- Increase in % of constructive edits (non-reverted) on mobile web
- Enable non-editing content contributions by mobile editors, such as highlighting confusing sections or identifying unreferenced material
Increase the number of retained editors who contribute primarily or exclusively on mobile devices. In particular, we hope to prioritize an increase in the number of new editors using mobile devices (with any increase in the number of existing users who edit more on mobile being a desirable effect of our work, but not an explicit priority)
Mitigate (baseline success) or reduce (strong success) the burden on experienced editors and administrators as a result of the increase in new editors using mobile devices

Additional measurements:

We expect to see improvements in qualitative/survey metrics across various personas (new mobile editors, organizers, experienced editors, administrators, etc.):
- Satisfaction
- Accessibility
- Campaign success for local organizers
We should also track downstream effects of these changes, without necessarily defining them as success metrics. These could include article quality, reference coverage, administrator satisfaction, diversity among editors (across various axes), edit "survival" rate (reduced reversion of content)