A lightweight lyric creation tool for unlocking small sparks of creativity
Stitch
A story-first vlogging tool that helps everyday people turn scattered clips, photos, and videos into meaningful narratives - without overwhelming them
Project Duration
8 weeks
Industry
Video creation
Creator productivity
AI-assisted storytelling
Team
Tools
Amy La (Designer)
Figma
Notion
Protopie
Justin Kim (Designer)
Piper Yu (Designer)
CONTEXT
People watch billions of hours of video, but rarely make any themselves
People consume massive amounts of media
YoutTube alone sees 5 billion hours of watch time every day, and a huge percentage of that is vlogs - simple videos documenting someone’s life.
The paradox: If we love watching other people’s everyday stories, then why do so few of us create videos about our own lives?
Our hypothesis
Could we build a tool that helps more people create vlogs that are: easy to make, meaningful to them, and genuinely fun to watch?
RESEARCH
Editing tools overwhelm beginners
So, we audited existing beginner video creation, advanced video creation, and general AI tools.
Beginner video creation
AI tools
Pro video creation
TikTok hides the timeline, Instagram’s template copies clip length, and iMovie’s vertical timeline give focus.
User interviews
“I take so many videos, but when I sit down to edit, I get stuck. I don’t even know where to begin.”
“I’m confused by the tools. I don’t know what the easiest one is to start with, so I just don’t start.”
Our takeaways
Beginners don’t know the story they’re trying to tell.
DESIGN PRINCIPLES
Designing for people who are just getting started
Guidelines that help reduce friction, build confidence, and make storytelling feel intuitive.
Fail forward
Mistakes shouldn’t break momentum. Like a map rerouting, the product should adapt when things go off-path.
Center authenticity
The best stories are real. Our job isn’t to fabricate fake moments. It’s to help shape what actually happened.
Creators stick with the process when they can see the destination. We need to reveal the outline early.
Make the vision visible
Our mission moved away from making editing simpler to make daily storytelling feel natural.
The shift
SOLUTION
A Walkthrough of a trip to Napa Valley
A real story. Amy captured clips on the go, but didn’t assemble the story until she was back home.
User adds a note about the itinerary before the trip
Last Edited 05-04-25
Me. My dad. And Alaska.
Running from a bear
Your projects
New project
Last Edited 08-20-25
Saying goodbye at our final retreat
Crying when I had to leave
Choosing a framing option
User has downtime and adds media to the brainstorm page
Adding more media and context after the initial script
STORY FINDING WITH AI
If editing isn’t the bottleneck, story discovery is
Our tool needed to help users find, shape, and express their story.
LLMs became central. The prompt box became a canvas, not a command.
Redefining the input
A textbox leads to generic prompts (“Make a video about Napa Valley”). With more context and media, the output becomes far richer.
This insight shaped the brainstorm page: a space that encourages users to drop everything: clips, photos, voice notes, thoughts.
Maximizing context (why audio matters)
People speak differently than they write. They ramble, reflect, and reveal emotional beats they’d never type.
So, we added a large, central record button. Spoken context helped our model understand not just what happened, but also why it mattered.
“I DON’T HAVE ENOUGH MEDIA”
Removing the pressure to have the “perfect” media
A lack of media should never be a blocker.
Rethinking what a vlog is
Our first approach was shot templates. But, real life can’t be recreated.
The essence of a vlog isn’t perfect footage. It’s the creator and their story. Even a simple talking-head clip can carry a narrative.
Making talking to the camera less scary
Teleprompter recording became core. At first, our early prototypes had a bold “Record now!” alerts, but this tested as emotionally stressful.
We asked a different question: What if the barrier isn’t remembering to record but feeling comfortable doing it?
Recording became modular
Users can record and edit in chunks (chapter by chapter), voice-over when visuals are not enough, and pause/resume anytime.
RETHINKING THE TIMELINE
Traditional timelines assume expertise
A blank timeline kills momentum.
What if editing didn’t involve dragging clips at all?
What if you edited the story instead?
Helping users see their story at a glance
We broke vlogs into chapters: title, media count, and duration.
This created a “story map,” showing which parts need more footage, where pacing dips, and where emotional peaks should be.
Giving users control without complexity
There are 2 layers of editing.
Global controls control tone, pacing, and direction, while local controls control chapters, sentences, and media.
At the end of the day, recording is available anywhere
and brainstorming is always one tap away.
RESULTS
A few key takeaways
Comfort matters
Recording issues weren’t about forgetting. The real barrier was emotional - feeling comfortable talking to the camera made recording easier.
Story drives everything
Beginners don’t need the perfect footage. They need clarity. Once users understand their vlog arc, they were far more likely to finish it.
Prototype to learn, not to prove
Early concepts like rigid shot templates revealed what didn’t work. Tight loops helped us quickly move toward natural, forgiving flows.
Build around real, not ideal behavior
Instead of designing for perfect filming scenarios, we built for reality: scattered clips, missed moments, and spontaneous ideas.
Reach out if you want to see earlier explorations, alternate flows, or the thinking that didn’t make the final cut.
ADDITIONAL EXPLORATIONS
Plenty more lives in the design archive
2025 @ Maya Parthasarathy
X