Page MenuHomePhabricator

Add an image: Conduct usability testing on Iteration 1 of the Add an image task
Closed, ResolvedPublic


Conduct user testing on designs proposed for the mobile-only Iteration 1 version of the "Add an image" structured task.

Multiple foundation teams–Android, Growth, Structured Data, CPT, Research–have collaborated together since October 2020 in exploring the potential of an image suggestion feature for Wikipedia articles.
After promising results from initial phases of exploration, the Growth team is building the first iteration of the feature (“Iteration 1”) which will show suggestions of images to add to unillustrated Wikipedia articles to Growth’s audience of newcomers on the mobile web.

Two distinct design directions were developed that explored the tension between providing newcomers with a focused and contextually-relevant task UI, with a design more similar to the Visual Editor for perceived greater on-ramp to the types of non-structured task editing.

  • A. “VE-Lite” - concept A hews more closely to the Visual editor UI, in that the article content appears more interactive and editable similar to how it would be in VE mode. The hypothesis is that this can help users ‘ramp up’ and familiarise with editing in an unstructured context.
  • B. “Task-focused” - concept B focuses the user on the task at hand, and centers the elements required on each step. The hypothesis is that providing clear focused prompts in context that are relevant to each part of the task (evaluating image vs writing a caption) will help users to succeed in this specific add an image task.

Target audience
The focus for this test will be those with limited experience editing on Wikipedia or Commons on mobile in the Growth team’s target markets (LatAm and MENA) for Iteration 1 release.

  • Experience contributing: Few or no edits to Wikipedia or Commons.
  • Platform: Mobile.
  • Languages:
    • English - due to resource constraints;
    • Spanish - ideally participants located in South America.

Primary objectives
By reviewing how well people are able to complete the task of evaluating an image and writing a caption using the two different approaches, we can decide on the overall leading design approach to iterate on as the final design for launch as “Iteration 1”, as well as identifying opportunities for usability improvements throughout the design.

Secondary objective
Gauge the general perceptions and attitudes about the task of evaluating image suggestions as a form of contribution on Wikipedia.

Test format
This will be an unmoderated remote task-based test, using to recruit participants and record sessions in English and Spanish.


  • Research plan:
  • Create prototypes on which to run the tests (subtasked as T287904)
  • Set up test in and run 1 pilot
  • Review and finalise test protocol accordingly
  • Run tests for Concept A and B with english participants
  • Translate the finalised English user testing protocol into Spanish
  • Set up Spanish test in and run 1 pilot
  • Run tests for Concept A and B with Spanish participants
Updated with key findings presentation

Related Objects

Resolved mewoph
Resolved mewoph
Resolved mewoph
Resolved mewoph
Resolved mewoph
Resolved mewoph
Resolved mewoph
Resolved mewoph

Event Timeline

kostajh triaged this task as Medium priority.Sep 10 2021, 12:43 PM
kostajh added subscribers: MRaishWMF, kostajh.

@MRaishWMF should this be in QA or another column?

@kostajh I believe QA is appropriate for this, given that all TODO's above have been completed.

Marking as Resolved because I don't believe there is call for more engineering QA on this task.