How2To
Step-by-step tutorials that sit in the room with you.
- My part
- UI design and UX research, and leading the sprints that turned Figma into Unity assets
- When
- 4 months, at Omnia
- Built on
- visionOS, Meta Quest and iOS
Research Methods
- 01
I held five internal sessions to ideate onboarding flows, spatial interactions, and modular templates like stove markers and screw guides.
Tap to flipFull notes for Brainstorming Sessions:
Held 5 internal sessions to ideate onboarding flows, spatial interactions, and modular templates (e.g., stove/oven marker for cooking, screw guide for repair).
- 02
I analyzed platforms like JigSpace and WikiHow VR, finding gaps in interactive guidance, spatial fidelity, and cross-context templates.
Tap to flipFull notes for Competitor Analysis:
Analyzed spatial learning and tutorial platforms (e.g., JigSpace, WikiHow VR, YouTube with AR extensions).
Identified gaps in interactive guidance, spatial UX fidelity, and lack of cross-context template systems for different task types (DIY, cooking, tech repair).
- 03
I interviewed nine participants who used online tutorials, finding most struggled to switch between tasks and wanted hands-free instruction.
Tap to flipFull notes for User Interviews:
Interviewed 9 participants (ages 20–45) who regularly engage with instructional content online (YouTube, blogs, mobile apps).
- 7/9 struggle to pause / switch between content and task
- 6/9 expressed interest in hands-free visual instruction
- 5/9 found current AR tools either non-intuitive or overly complex
- 04
I built three user personas from interview data: a DIY repairer, a home cook, and a computer builder.
Tap to flipFull notes for Persona Building:
Constructed 3 personas from interview data:
- DIY Dan: 32, uses YouTube for repairs, needs visual clarity
- Chef Lila: 28, uses Instagram Reels for recipes, wants step control
- Tinker Tim: 24, builds PCs, frustrated with switching apps during work
- 05
I clustered interview responses around multitasking load, switching friction, demand for spatial tutorials, and headset interest for hands-free guidance.
Tap to flipFull notes for Affinity Mapping:
Clustered responses into core themes:
- Cognitive load during multitasking
- Friction in switching between instruction and action
- Demand for real-time, spatial, context-aware tutorials
- Excitement around using headsets for hands-free guidance
Show full notes
01Brainstorming Sessions
Held 5 internal sessions to ideate onboarding flows, spatial interactions, and modular templates (e.g., stove/oven marker for cooking, screw guide for repair).
02Competitor Analysis
Analyzed spatial learning and tutorial platforms (e.g., JigSpace, WikiHow VR, YouTube with AR extensions).
Identified gaps in interactive guidance, spatial UX fidelity, and lack of cross-context template systems for different task types (DIY, cooking, tech repair).
03User Interviews
Interviewed 9 participants (ages 20–45) who regularly engage with instructional content online (YouTube, blogs, mobile apps).
- 7/9 struggle to pause / switch between content and task
- 6/9 expressed interest in hands-free visual instruction
- 5/9 found current AR tools either non-intuitive or overly complex
04Persona Building
Constructed 3 personas from interview data:
- DIY Dan: 32, uses YouTube for repairs, needs visual clarity
- Chef Lila: 28, uses Instagram Reels for recipes, wants step control
- Tinker Tim: 24, builds PCs, frustrated with switching apps during work
05Affinity Mapping
Clustered responses into core themes:
- Cognitive load during multitasking
- Friction in switching between instruction and action
- Demand for real-time, spatial, context-aware tutorials
- Excitement around using headsets for hands-free guidance
Voice-first UI
Unobtrusive Design
I designed a voice-first interaction model that eliminates cognitive overload by streamlining commands to only the most contextually relevant actions. On Vision Pro, this voice layer acts as a spatial tutorial generation assistant, intelligently creating steps enriched with videos and images that are task-specific, enabling users to focus entirely on hands-on activity. The design follows Minimalism principles and prioritizes flow continuity in 3D space.
Empathetic Design
Tutorial Preview Page
Once the AI generates a tutorial, this page allows the user to go over the results, sources, and ingredients. Once everything is in order, they can proceed to the immersive viewer. This page enables options and a means of knowing if they have all the right tools/ingredients to perform the task they asked for. If they don't approve of the tutorial, they can head to the 'Browse' page for others, 'See More', or the Voice search to regenerate with a prompt, or 'Continue' to proceed to completing the task.
Tutorial Views
Immersive + Condensed Views
Designed two tutorial viewer sizes to suit all user needs. They can choose to be fully immersed to stay focused on a single task or take a better look at the step. The condensed view would allow more information and multi-app usage. From user testing, we found that users who are accustomed to spatial interfaces tend to have 3-4 applications running in their view, and the condensed view is best catered to them. Similarly, it is highly demanded in B2B settings where worker efficiency is preferred over content immersion.
Tutorial Gallery
Expansive Gallery view
My work focused on creating an intuitive and aesthetically pleasing user experience. I prioritized clarity and accessibility by organizing content into easily navigable sections, such as “Food”, "DIY", "Technology", etc., using distinct, colorful cards to highlight generated tutorials. This design not only enhances user engagement but also ensures that information is both discoverable and enjoyable to browse.
Video Controls
Reimagined Media Control
Due to the extensive functionalities of How2, I redesigned the native media controls. Now we have Steps, access to tutorials, a toggle to change the viewer mode, show or hide ingredients, and even information sources (described from left to right). This design maintains a small footprint and ensures spatial efficiency without sacrificing functionality.
