Quora connect the people who have knowledge to the people who need it, to bring together people with different perspectives so they can understand each other better, and to empower everyone to share their knowledge for the benefit of the rest of the world.
As a huge fan of the product, I cared very much about the user experience. The Browse and Search Results mobile Android features bothered me, so I conducted a moderated usability test with 5 users to gather data about the understandability of the features, and presented recommendations.
Identify the pain points of Quora’s current “Browse” and “Search Results” interfaces:
- WHAT: Quora Android Mobile App
- WHO: Existing Quora users
- WHY: To test whether others experience the same usability issues I’ve identified in Quora’s Browse and Search Results interfaces.
I sat down with 5 individuals and captured notes while they performed the following tasks:
- Homepage Tour
- Look for Topic-Specific Questions & Answers
- Search for a Topic
- Browse for Content of Interest to the User
Tasks were determined to test my hypothesis that Quora’s Browse and Search Results need improvement.
To prevent leading questions and prescriptive tasks, I framed the tasks in the form of scenarios:
“I’d like you to imagine that you are really interested in a topic such as design, dogs, or Miley Cyrus, but don’t have a particular question in mind. Being curious and not having a lot of time, you want to find all the top questions and answers in the topic of your choice. Show me what you would do to solve your problem.”
Findings: Two Key Issues
Users expect interesting content recommendations (but get Unanswered Questions, People to Follow, and Shuffle).
“This is terrible. I expected to see things that may be interesting. Doesn’t seem like a good way for people to actually discover new stuff. ”
Upon clicking the Browse page, users were confused because they expected a screen for personalized content discovery. “Unanswered Questions” is geared more towards the 1% of users in the oft-quoted 1-9-90 rule (1% will answer, 9% will edit, and 90% lurk) in content-based networks. “People to Follow” is one degree of separation from from the personalized content that users want (i.e. Questions / Answers, Topics). “Shuffle” is too undirected and doesn’t cater to the user’s interests.
Currently, users are introduced to the Browse feature is by reading through the “I Just joined Quora from my Android phone or iPhone. How do I get started?” question, which is shown after a user signs up. But as the test points out, users are unclear why “Unanswered Questions”, “People to Follow” and “Shuffle” belong in the Browse page.
ASSUMPTIONS: Although all 5 users found the features in the Browse page confusing and questionable, there is no statistical significance to generalize these behaviors / preferences across all of Quora’s users. However, with roughly 90% of users lurking, there’s still a high probability that other users face the same challenges.
RECOMMENDATION: Allow users to determine what “interesting” content means to them for personalized content discovery, as they expect to. Redesign the Browse page to allow self-conducted curation and customization of content. Self-curated collections is a prevalent design pattern within content-based networks (i.e. Medium, Pinterest, Youtube).
Users find it difficult to sort through search results because they all look the same and because the quality signals are hard to find (i.e. upvotes, number of answers, answerer background).
“All the results look the same — no distinction between topic and questions. I had to sort through unwanted stuff (blog, questions, answers) just to find topics.”
RECOMMENDATION: Allow users to easily distinguish between the different search result types and see clearer quality signals. Improve the visual hierarchy for search results by making the differences between search result types more apparent (i.e. Topic vs. Blog vs. Profile) and by displaying highly-visible quality signals (i.e. Upvotes). Possible solutions include typography adjustments, search results filtering, and other applications of Gestalt Laws.