uWave is a lifestyle and community platform for students to discover local and interest based content.
Our initial feed was limited to university community. As we opened up to more users outside university, the feed needed to be designed for general gen-Z users and their interests.
More users and generalised content led to:
More diverse content: this made it harder for users to navigate relevant content based on their interests
Larger content quantity: more content quantity made is difficult for the users to tune their feeds
User interviews
One-on-one user interviews with our power users to gain insights over the problems users were facing.
Usability testing
Conducted in-person as well as online user testing with the MVP. Observed users as they interacted with the prototypes to identify areas where they were struggled
Our competitors
Target Users
uWave targets the Singapore market and aims to cater to students and gen-z users looking to find good quality content and authentic reviews as well as make social connections in the digital world.
User persona
USER INTERVIEWS
User painpoints
We spoke to and interviewed multiple users of social media apps. And we found out this:
More users and generalised content led to:
Lack of authenticity and trust: Nowadays, social media is filled with influencers doing sponsored or paid reviews, which leads to a lack of trust and authenticity.
Weak Local Features: Most social media apps lack filtering capabilities to find local, relatable content.
Lack of Emphasis on Interest-Based Communities: Currently, social media apps do not place much emphasis on bringing users together around common interests.
FROM USER INTERVIEWS
Observations
We spoke to and interviewed multiple users using our app. And we found out more users and generalised content led to:
More irrelevant content
Users were spending time on the platform, but the engagement dropped. This meant users were spending more time in finding relevant and interesting content.Difficult to discover content
Users were finding it difficult to tune their feeds, as more users were creating content for different interests, and not limited to university issues and discussions.
Key goals
Feed Design
How may we help users seamlessly browse content based on the their interests and local community inspired.
Internal tagging system
How may we enable internal users to efficiently label and moderate the data, so that end users can easily find more relevant content.
Proposed Solutions
Interest based discovery experience
Allow users to easily tune their experience based on specific interests such as Cafe, Books, Fitness, Adventures etc.
Users should be easily able to navigate to different interest based sections.
Network-based and interest based feed
Provide users with an opportunity to explore diverse content along with the ability to shape their feed. We introduced two feed sections for users:
Network-based feed: Allow users to follow specific creators and communities to shape their feed based on their network.
Interest based feed: Allow users to personalize their feed based on specific interests. This is a place for users to explore and discover content from a wider community, and not limited to who they follow or which community they are part of.
Home Feed
Home feed is a network based feed, that allows users to browse and engage with content based on the following and communities.
Explore Feed
Explore feed is a interest based feed, that allows users to discover a wider range of content from more creators in a personalised way.
Search page with keyword directives
Other feed improvements
Ability to create and consume different types of content in a mixed feed format, including text, image, video content, and polls.
Reactions allow Gen-Z users to express themselves and improve engagement with the content.
Internal Content Tagging System (web)
The internal content tagging platform is a system designed for manually tagging the relevant content to improve the search results and recommendations for our users.
The internal content tagging system is designed to enable efficient tagging of content. It helped reduce the number of clicks to save time for faster tagging.
FROM USER INTERVIEWS
Observations
Incorrect category selection by the users
Self-selection of content categories by the users is prone to incorrect and inconsistent categorisation of content. This leads to irrelevant content recommendation to the users.
A mandatory category selection step added friction to the users
Making the category selection mandatory before posting any content added friction to the users. We observed that many of the users stopped at this step while posting the content.
Proposed Solutions
Internal tagging system
Develop a system to easily categorise, tag and rate the content in order to improve recommendation quality.
The tagging system can also be used to flag content that is not aligned with the platform guidelines such as violence, nudity etc.
The labeled data can be used to train machine learning based recommender system in future to further improve content recommendation.
Remove the mandatory step for the user to content category
We proposed to remove the mandatory step for selecting a category for content to reduce the friction for users.
Internal content tagging system- After the content is tagged
Impact & outcome
