Wikimedia Foundation Annual Plan/Quarterly check-ins/Audiences2 Notes October 2018

Notes from the Quarterly Review meeting with the Wikimedia Foundation's Audiences teams, October 31, 2018, 8:00am - 9:30am PDT.

Please keep in mind that these minutes are mostly a rough paraphrase of what was said at the meeting, rather than a source of authoritative information. Consider referring to the presentation slides and other official material

QCI Audiences #2 session

Intro to new content programs: Toby

TN: "all models are wrong, but some are useful" - this one is obviously incompletely, e.g. doesn't contain donors
We call it a flywheel cos each aspect supports each other
At the moment we are still in research and design phase
[chat question: what is new content about? https://meta.wikimedia.org/wiki/Wikimedia_Foundation_Annual_Plan/2018-2019/Audiences#New_Content ]

Progressive onboarding

Danny: found that retention is really low
need tools to have newbies learn abotu practices etc

Abbey: research project on new editors' needs
mapped those needs to personas

Growth Projects: Marshall

is about increasing retention of new editors
retention is a problem especially in mid-size wikipedias
funnel for kowiki – ~45% ever make an edit, ~8% come back to make another edit; seek to increase both of those proportions
built on new editors research by abbey et al
focus on day 1 - most who make edit do it on day 1
3 ideas selected from a long list based on in-team discussion, research and Community consultations

personalized 1st day: don't know so far what [aim] people arrive with
in-context help
all these 3 projects in flux currently
depending on initial outcomes may also start pursuing a couple of other top design interventions (Engagement emails and in-context lessons/tutorials in Q3)

Mobile Contribution

Danny: lots of new internet users in middle income countries where laptops are expensive but everyone has a phone

Abbey: mobile personas in US
... in India
Margeigh on mobile contributions learnings
Toby: understanding users in their context then using personas to make this influence development

Contribution tools on Mobile Web: Olga

Olga: Mobile Web improvements for existing contirbutors
Began with design research (interviews, data, feedback on prototypes)
design looks quite busy, on purpose - added most links that editors have on desktop
Toby: Still going to have the floating toolbar??
Olga: Still tbc, but there will be sticky top bar, more in side nav
Olga: Initially Editors are super excited about the work, learnings through research as well about how to reach out to and communicate with Editors for feedback
Olga: team made mental switch from supporting readers to editors. some of these had been requested for a long time
Toby: very exciting
Olga: ... (next steps)
Toby: moderation tools are just important as editing tools, so it's great to see work on this aspect (curation, e.g. making Special pages more usable and accessible on mobile)

VE on mobile: Dan

this q: find out what challenges people face, how we can address them
Generally VE edit success was lower than Wikitext since more experienced users used Wikitext, but breaking down by all experience levels, VE actually does perform better.
Another key finding from Heuristic research was that help was missing in mobile VE - confirmed previous NE research and work being done on Growth
making VE faster should make fewer people drop off

Mobile for existing contributors on iOS: Josh

Josh: this has been a design and research quarter
added wikidata description editing (which Android did last year) - not very tightly aligned, but good to get start on contribution
Chelsy provided great data research report on mobile app editors on iOS - first time done in the apps and will inform future metrics
A ton of design research (generative, evaluative) by CMadeo to review other mobile editing contexts as iOS ramps up work on contributor UIs
e.g. notetaking and coding apps
Toby: if you have questions or concerns whether people contribute on mobile at all, read that report: they do
...
Josh: heard from editors that navigating within wikitext can be a challenge, so we're designing some features to help with that, starting with syntax highlighting
data: as jk discussed in yesterday's QCI, error bars are like 90% in some of the metrics [that were initially envisaged in the AP], so need to find more suitable ones

Mobile for existing contributors on Android: Charlotte

notifications: released on Monday
can we move the app editor retention metric with these initial Welcome, Milestone and Thanks Echo notifications in the app?
V0 test but also support longer term goals for introducing editor Uis into the apps

Small mobile contributions will be introduced next quarter
- intentionally not showing mocks yet, just want to introduce the concept of how users can contribute simply in smaller bites
- there will be "quality gates" on each contribution type being trialled to ensure these small tasks are not [damaging] - like last year the revert rate for Wikidata description edits
Toby: even though there is a large difference in (readers) audience size between apps and web, apps can become a fully valuable platform for contribution at current dimensions already

Content Translation: Pau

V2 of Content translation (CX) tool will be slowly introduced after initial rounds of feedback, with end goal of making V2 the default
Improvements for Translate extension, connected to TranslateWiki.net (Pau)

Release Cadenece: Toby

Toby: there were questions about release cadence

Jaime: how are you going to use model going forward?
Toby: understand how pieces fit together, how they inform team
e.g. when previously focused on readers
Jaime: model is useful, helps rest of us understand what you're doing
Toby: also helps to inform outreach efforts, for knowledge equity
JK: in the past struggled with funnel model - community is complicated, refine
Josh: might even quantify some of this