StudyBuddy
(2024)
Duration
3 Months
Role
Sole Product Designer
Project
Studybuddy is a co-studying platform that match users for 1-on-1 online sessions.
A YC-backed started now serving 12,000+ user to not have to study alone
StudyBuddy matches you with a partner so you can stay accountable and focused. After joining Y-Combinator in 2023 the platform has grown to 12,000+ users worldwide, building a community of people who support each other in staying focused.
It All Started With A Business Problem
Loosing users after first-time user experience (FTUE)
With the launch of the subscription membership model, the high drop-out rate became a crucial gap and presents a major business opportunity: improving the first experience to increase the likelihood of users converting to paying members.
Complicated feeling towards meeting a stranger
After conducting mixed-methods research—including semi-structured interviews (15 users), competitive analysis (5 products), and a survey (60 responses)—to explore users' first experiences on StudyBuddy, I uncovered a common theme: anxiety around interacting with a stranger.
I identified 3 critical moments where first-time users feel overwhelmed. And throughout all the moments, anxiety around macthing, meeting and is the key reason why users drop off.
I arrived at this design question to help my design process. The goal was to help ensure users have a smooth first-time experience.
Address Drop-Out Friction Points Using Design
With a low user base on the platform, users are experiencing long wait times for a match. This delay impacts the overall user experience and could lead to frustration or drop-offs.
I explored how other products are addressing the inevitable wait problem and came up with three directions.
I transformed a system limitation—idle wait time—into a user-centered opportunity by applying Behavioral Momentum. Introducing a low-friction micro-action during this period helped users engage quickly, turning uncertainty into purposeful activation and reducing drop-off.
Before
Users lacked visibility into the system's status, leading to anxiety about whether they would be matched. Once matched, they often felt unprepared to enter the meeting room. These moments of uncertainty and cognitive friction contributed to early drop-off.
After
An animated facilitator now provides a sense of presence and guidance during the wait time, prompting users to begin their first task and making the wait feel purposeful. Simultaneously, real-time system feedback offers transparency and reassurance, reducing uncertainty and increasing user confidence—ultimately lowering the risk of drop-off.
Adding a facilitator role to ease the stranger anxiety
Design a facilitator animation to engage users while they wait, guiding them through the process and nudging them to start their first to-do list.
Meanwhile, provide real-time system updates to keep them informed. This approach helps reduce anxiety and the likelihood of drop-off, creating a more seamless and engaging experience.
Meanwhile, provide real-time system updates to keep them informed. This approach helps reduce anxiety and the likelihood of drop-off, creating a more seamless and engaging experience.
Final design - home & matching
The start and end of sessions can feel awkward, especially with greetings and deciding whether to extend the session. These moments of uncertainty can make users feel uncomfortable and disengaged.
I explored 3 AI-driven solutions to act as facilitators in the study room, aiming to ease the social awkwardness users may feel when meeting a stranger. The key nuance was balancing the level of AI automation and presence—ensuring it supported the experience without distracting from the meeting or study focus.
After evaluating the options, we chose a scripted chat approach to reduce technical friction and minimize distractions.
Before
Users lacked guidance or design cues to help initiate conversation and break the ice in the study room.
After
With the scripted chatbot, users can naturally break the ice through facilitator prompts in the chat. The toolbox in the top-left corner also supports non-verbal collaboration, helping users make mutual decisions without needing to talk.
Final design - study session
Without feedback after a session, users often expect the same experience in future sessions, even if the previous one didn’t go perfectly.
Early preferences are often unstable or based on assumptions. so it’s important to design a deliberate UX moment after the first experience that checks in and asks: “Now that you’ve tried it, want to adjust your setup/preferences?”
Add a light preference check at the end of each session where the AI facilitator assesses the vibe of the session. This gives users control over their preferences for future matches, helping them feel more confident about improving their experience going forward. It also ensures a positive closure to their first experience, setting expectations for future sessions.
Before
Users are directed back to the home page after the session ends. There was no feedback or wrap up. Users lack of closure and opportunity to alter expecations so they have no confidence to come back.
After
With a chance to alter preferences and a sharable stats, users can alter their preferences and feeling more prepared to start their next session. The shareable stats also serve as a rewarding experience for users to feel accomplished after a session.
Final design - post-session check-in flow
Final Design, Up Close
What I Took Away From the Design Journey Is...
Early drop-off = wasted acquisition cost. A well-executed recovery can actually improve brand loyalty and trust — people appreciate when a company notices and cares.
It's all about detecting and repairing a bad or incomplete experience before it leads to user churn or dissatisfaction.
It's all about detecting and repairing a bad or incomplete experience before it leads to user churn or dissatisfaction.
What Changed with the New Design?
70% Drop-off
After implementing onboarding tutorial flow and partially implementing the meeting view design, drop-off rate decreased 70%.
120% SUS Score
In a usability test with 5 users, the SUS usability score of the system improved by 120%, leading to a better user experience.
