Class project
Bumble:
Buzzcalls
Desc.
Designing a novel premium feature where users can join short, AI-moderated video/voice chats with promising potential matches to form real connections.
role
Product Designer
Team
Solo
timeline
4 weeks
skills
UX Design
User Testing
User Research
Prototyping
overview
problem
solution
research
user flow
proposed kpis
reflection
Overview
Bumble is a popular dating app where women make the first move in heterosexual matches. Bumble has developed a high degree of trust for the 50 million active users, the vast majority of which do not pay for the service.
The Bumble Customer Acquisition Team aims to increase the conversion rate of current non-paying users. It also aims to attract new users in all categories, particularly users in “BFF” and “Bizz” categories who are not dating.
problem + user research insights
Likes are meaningless if people suck at making profiles and texting.
“Any conversation that starts with a dead ‘Hi’ is immediately low-momentum and low-energy, it's hard for either side to get excited. [...] Out of frustration, I eventually started saying ‘Hi’ back and then it would instantly die out.” -Reddit User
Meaningless matches. Dying conversations. Boring profiles that realistically show nothing about a person. These problems lead to many users deleting the app out of frustration; premium features don’t address the fundamental flaws that come with static dating profiles. Learn more about my user research below.
solution
BuzzCalls: A video-based speed dating function.
01 onboarding + timed cycle integration
Switching from normal scrolling view to BuzzCall mode; Camera + mic check; Onboarding cards.
02 Profile review + match
Once the user has entered the BuzzCall cycle, a timed profile review begins.
03 Camera check and enter call
When both users swipe yes, the user begins a final camera/mic check before entering the call.
After entering call, users can choose from prompts at the bottom to discuss.
04 Time up and match
Time's up! User decide if they want to match or not.
other features
Notifications are sent to the user when more people join the BuzzCalls feature to minimize wait time.
research
Trying to find the core problem to fix was the most difficult part. To do this, I had to do some "market research" myself: by downloading Bumble.
To perform the most immersive research, I downloaded Bumble to test it out myself. I thought that maybe it'd add some excitement into my single life, and it did for a while, but nothing really went anywhere.
A common problem I noticed was that people are really bad at setting up their profiles. Like REALLY bad. Matches were common because sending likes are easy. Even if we matched by sending likes to each other, there was no real starting point for conversation. I don't know what to say besides "hi", which defeats the whole purpose of the "woman first" thing.
The problem I decided I have to focus on solving was getting past that first stage of awkward static friction where you don't really know how to approach an interaction.
Examples of some very….interesting profiles. 1 chooses not to use any text prompts, only puts one photo, and includes very minimal information. 2 is a creative catfish profile. 3 is a tragic example of not knowing how to present the full potential of oneself through photographs.
insights
Texting doesn't seem like the best way to approach finding love.
Research comparing text, audio, and video communication shows that richer modalities, especially video, lead to higher levels of liking and perceived closeness during initial interactions than text alone. This supports BuzzCalls as a powerful feature for Bumble: by prompting early video conversations, BuzzCalls reduces meaningless matches, dying chats, and reliance on shallow profile signals. Video calls surface real personality cues such as tone, humor, and chemistry, enabling users to assess compatibility faster and more authentically than prolonged texting. (ScienceDirect)
reframing + proposed solution
I reframed the problem around what I saw as the real issue: people struggling to form genuine connections through static profiles and small talk.
By focusing on real-time interactions like video calls, I shifted the idea toward helping users better express their personalities and build real, authentic connections. This direction creates more value for both users and the organization: users can have more meaningful experiences on the app, and Bumble benefits from higher engagement and potential conversion rates through deeper emotional investment and success stories.
user flow + revenue touchpoints
proposed key performance indicators
reflection
Solving the right problem will almost always outweigh the importance of adhering exactly to a prewritten design brief. Overall, the golden spot is finding a solution that creates real value for users and the company, balancing company requirements and user needs.
To solve the right problem, you have to think outside the hive mind :) 🐝
© 2025 Cindy Tso
cindy.tso05@gmail.com