Pinkoi Discover
Shopping Doesn't Always Start with Search
0→1 · Product Discovery · Growth
Role:
Lead Product Designer
Design:
2024/09
Rollout:
2024/11
Platform:
App·iOS & Android
01
OVERVIEW
Shopping Doesn't Always Start with Search
02
RESEARCH
Start with Research to Define What Discovery Should Be
01
Users paid close attention to design styles and trends, with imagery as their primary medium for discovering new ideas
02
The opportunity was not to add more content, but to make existing content easier to explore and understand
03
Users with different style preferences showed no meaningful differences in their information sources or content needs
03
GOAL
How Did We Define Success for Discover?
Discover was not designed to drive an immediate purchase. Its role was to support the exploration stage—helping users without a specific goal find inspiration and gradually build interest in content and products.
We therefore defined success through two key behaviors:
Goal 01 Conversion to Product
Move from Content to Product Discovery
Measures whether Discover turns inspiration into intent to explore products
Discover → Product Page Click-Through Rate
Product Browsing Depth
Goal 02 Continued Exploration
Keep Discovery Going
Measures whether users continue exploring instead of leaving after one piece of content
Theme Browsing Depth
Engagement such as saves and shares
04
DESIGN STRATEGY
Using Design to Explore Three Product Directions
Before the product format was defined, I translated research insights into three exploration models. Interface prototypes helped the team compare each direction's experience, constraints, and scalability
Focus on one theme at a time and keep discovering through swiping
A full-screen experience focused on one theme, with horizontal and vertical swiping for continuous exploration
Strengths
Strong visual impact and a high sense of discovery
Trade-offs
More complex interaction model and greater reliance on high-quality assets
Scan multiple themes quickly, then dive into what interests you
A feed of imagery and summaries enables fast scanning before users open the themes that interest them
Strengths
Easy to understand, efficient to browse, and better suited to high-volume multilingual content
Trade-offs
Less immersive, with fewer moments of interaction delight
Lead with imagery, then invite deeper exploration
Strengths
Aligns with the research insight that imagery is a key source of inspiration
Trade-offs
Weaker content context and stronger dependence on image quality
Lead with imagery and use copy as support, making visuals the primary entry point to discovery
Design layouts that adapt to different languages, text lengths, and localized content
Keep the content architecture and interactions extensible for future taxonomy and recommendation systems
05
FINAL DESIGN
Designing Discovery as a Rhythmic, Extensible Experience
01
Subtle motion across list and detail pages helps users understand their position and perceive changes as they scroll, switch themes, and read.
List Page | Make Theme Changes Feel Clear
Card motion and transition feedback make the shift to the next theme clear
Detail Page | Use Motion to Guide Attention
Transitions across titles, tags, and content guide attention to key information
02
Product Discovery
Connect Content Naturally to Relevant Products
Related Content
Extend the Journey with Related Themes and Tags
03
Japanese Version
English Version
06
LAUNCH & ITERATION
Post-Launch Data Revealed the Critical Drop-Off
With limited engineering capacity, I focused on the primary drop-off rather than redesigning the entire experience, proposing four changes we could validate quickly:
Reduce hero image height
Bring the title into the first viewport sooner
Localize titles
Reduce the effort required to understand the content
Add content tags
Provide more context before users open a detail page
Create return cues
Use app push and NEW labels to signal fresh content
After the iteration, Discover's core metrics continued to improve:
PRODUCT IMPACT
Adoption Rate
Revisit Rate
The iteration showed that improving Discover was not about adding more content or redesigning everything. It was about communicating value earlier and lowering the barrier to exploration.
07
IMPACT
Focus on the Critical Drop-Off to Make Small Iterations Matter
01
Post-launch data showed that the problem was not the content itself; users were dropping off before entering a detail page. Focusing on this transition was more effective than redesigning the entire experience.
02
Reducing the hero image, localizing titles, and adding tags all helped communicate value earlier. After the iteration, detail-to-product click-through rate increased from 34% to 37.5%.
03
Discover's impact did not peak immediately after launch. As content accumulated and Push and NEW cues encouraged return visits, the return rate gradually climbed to 42.8%.
Recognition
08
FINAL DESIGN
09
TAKEAWAYS
How Design Drove a 0→1 Product Forward
01
In the ambiguity of a 0→1 product, I brought design into the decision-making process early, making different possibilities tangible so the team could clarify direction, evaluate trade-offs, and make decisions faster.
I quickly translated research insights into prototypes across different product directions, turning abstract discussions into tangible experiences that allowed the team to compare options, constraints, and trade-offs, while understanding the implications behind each interaction.
Rather than simply accelerating design execution, this made design a shared language for exploring product direction — helping the team build shared understanding, align quickly, and keep decisions moving forward even before requirements were fully defined.
02
After launch, I continued observing real user behavior, revisiting our initial assumptions and identifying the problems that actually had the greatest impact on the experience.
Rather than continuing to add features based on the original plan, I returned to user behavior to reassess the problem and priorities, focusing the next iteration on the highest-impact friction in the journey.
Launch is not simply the end of delivery, but the beginning of real signals. Using data to continuously refine judgment and decide what is most worth solving next has become an important part of how I drive product iteration.
02
Stepping beyond design to drive product iteration
When post-launch data revealed new opportunities, I stepped into parts of the Product Manager role — from interpreting data and framing priorities to working with engineers on feasibility, effort, and the next iteration.
Through close design–engineering collaboration, insights were quickly translated into product improvements, resulting in a significant uplift in key metrics.
This also reinforced how I contribute beyond design execution: identifying opportunities through data, making product judgments, and connecting teams to turn insights into measurable outcomes.
