The Problem
Photo-based matching creates shallow connections and high abandonment rates. Users invest time and emotional energy into conversations with people they feel no chemistry with upon meeting.
The Opportunity
Video technology is ubiquitous. The barrier to video is no longer technical — it's behavioral. The right UX can make video feel natural, not performative.
My Role
End-to-end UX design: research, synthesis, information architecture, interaction design, visual design, and usability testing.
How Might We
Design a dating experience where people connect with who someone actually is — not who they've chosen to appear to be?
Conducted 1:1 interviews with active dating app users aged 24–38 across Chicago, Denver, and San Francisco. Sessions focused on dating app habits, frustrations, and what "a good first connection" felt like.
Audited four major platforms — Tinder, Bumble, Hinge, and Coffee Meets Bagel — analyzing their matching mechanisms, profile structures, and the points at which users disengaged.
Distributed a screened survey to understand patterns at scale. Focused on frequency of use, reasons for abandonment, and attitudes toward video calls with strangers.
Across all interviews and survey responses, four themes emerged consistently.
"I never know if someone's photos actually look like them until we meet in person. By then I've already invested a lot of time."
Insight 01 · Authenticity Gap
"Swiping feels like scrolling through a catalog. I'll do it for 20 minutes and feel worse about myself than when I started."
Insight 02 · Swipe Fatigue
"The first date is always awkward because we've basically just been texting. I don't know how they carry themselves until I'm already sitting across from them."
Insight 03 · Chemistry is Physical
"I'd love a way to see someone in motion — just talking normally — before I decide. That would save everyone time and disappointment."
Insight 04 · Appetite for Video
Every major dating platform is built on the same assumption: a photo index is the best way to discover someone. None have challenged that model at the core experience level.
| Platform | Core Model | Strengths | Critical Gap |
|---|---|---|---|
| Tinder | Photo swipe, bilateral match | Large networkFast discovery | Gamification erodes meaningful intent. Photo-primary reduces personality signals. |
| Bumble | Photo swipe, women initiate | Safety focusIntent signaling | Same photo problem. Reduces harassment but doesn't improve authenticity. |
| Hinge | Curated prompts + photos | Personality promptsConversation hooks | Prompts are written, edited, curated — still performative, not genuine. |
| Coffee Meets Bagel | Daily curated match | Intentional paceQuality over qty | Limited pool, photo-primary, no mechanism for dynamic presence. |
| Vidate ↗ | Video-first profile, bilateral match | Authentic presencePre-match chemistryReduced deception | Opportunity: reduce camera anxiety, design for natural video self-expression. |
33 · Manager · Denver, CO · Single
Mark has used Tinder and Hinge on and off for three years. He's active, curious, and values genuine conversation — but finds the swipe loop draining. His best relationships started from meeting someone naturally, where he could "just feel the vibe." He's skeptical of apps because they've produced too many first dates where the chemistry just wasn't there.
App Usage
Daily, 15–30 min
Dating Goal
Long-term relationship
Video Comfort
Moderate — uses FaceTime regularly
Pain Points
Interviews, competitive audit, affinity mapping to identify the core tension: authenticity vs. performance anxiety.
HMW statements. Defined the design principles: Natural over Performative, Reduce Before Reveal, Chemistry First.
Sketched 3 concept directions. Validated the "short-form video profile" model over live-streaming and text-enhanced photos.
Built lo-fi flows in FigJam, then high-fidelity interactive prototype in Figma across 3 iterations.
3 rounds of moderated testing (6 participants each). Iterated on recording UI, discovery flow, and privacy controls.
Concept Storyboard · Early Sketches
[ Insert storyboard sketch image from Figma ]
Every major design choice was driven by a specific user insight. Here are the three decisions that shaped the core experience.
Recording UI Screen
[ Insert Figma export ]
Decision 01
Early concepts had no time limit. Testing revealed that open-ended recording created paralysis — users didn't know when to stop or what to say. A 60-second hard cap changed behavior entirely: people stopped scripting and started talking.
The recording UI uses a circular progress indicator rather than a countdown timer. Showing progress (not remaining time) reduced anxiety in testing by keeping attention on presence rather than deadline.
Design Rationale
Constraint creates freedom. The limit exists not to restrict expression, but to eliminate the performance anxiety of an open canvas. The format signals: be yourself, not a character.
Decision 02
Discovery was originally a full video play. Testing showed users skipped videos within 3 seconds anyway — but felt guilty doing it. The 5-second auto-preview model aligns the interaction with actual behavior: it's a glance, not an audition.
Users tap to watch the full video only after the preview lands. This created a two-stage attention model that felt more like real human interaction — and eliminated the guilt of "skipping" someone.
Design Rationale
Designing against user guilt was a core principle. If the interface creates negative emotion, it creates negative association with the platform — and by extension, the person they skipped.
Discovery Feed Screen
[ Insert Figma export ]
Speed Dating Session Screen
[ Insert Figma export ]
Decision 03
A spontaneous live video call between two strangers felt too high-stakes to users. The Speed Dating session model — pre-scheduled 5-minute live video rounds — reduced the cold-call anxiety significantly.
Users opt in to a session window. They're matched sequentially with a queue of compatible profiles for brief live calls. At the end of each round, both can "continue" or pass. No judgment, no awkward goodbye.
Design Rationale
Structure reduces risk. By gamifying the live session format, we gave users a clear start, middle, and end — removing the social friction of initiating and ending a video call with a stranger.
Three rounds of moderated usability testing with 6 participants each. Participants were screened as active dating app users, 24–38. Testing focused on recording comfort, discovery legibility, and trust in the platform.
Issue 01 · Round 1
5 of 6 participants paused before recording. Most said they weren't sure how long to speak or when to stop. The open format created a performance vacuum.
Fix Applied
Added the 60-second progress ring and a soft prompt overlay: "Just talk. Tell us one thing you love." Hesitation dropped significantly in Round 2.
Issue 02 · Round 1
Users described the full-video discovery feed as "watching YouTube," not dating. It shifted the interaction from personal to consumptive.
Fix Applied
Introduced the 5-second auto-preview model. Full video only plays on explicit tap. Reframed the discovery experience as "glancing" rather than "watching."
Issue 03 · Round 2
3 of 6 participants in Round 2 expressed concern about where their video would be visible. One declined to record until this was clarified.
Fix Applied
Added a persistent Privacy Control panel accessible from the recording screen. Users can set video visibility: Everyone / Mutual Likes Only / Hidden. Default set to Mutual Likes Only.
The final Vidate experience across four core screens: Sign In, Profile Discovery, Match, and Video Record.
Sign In
[ Figma export ]
01 · Onboarding
Discovery
[ Figma export ]
02 · Discover
Match Screen
[ Figma export ]
03 · It's a Match
Profile
[ Figma export ]
04 · Profile View