#SearchExperience #AIIntegration #ProductOptimization
Role:
Product Designer
Company:
Pinkoi
Year:
2026
Platform:
01
OVERVIEW
Aligning search results with shopping intent
OUTCOMES
02
CHALLENGE
More Search Results ≠ Finding the Right Products
01
Users may not know the exact product name or may search with broad terms. As a result, search results may not match their actual shopping needs, making it difficult to find the products they're looking for.
02
Existing search suggestions aren't always relevant to users' search terms. Some are repetitive or too similar, making it difficult for users to refine their searches or explore related products.
03
Different products have different shopping criteria, such as material, capacity, or intended use. However, fixed filters don't adapt to the search context, making it harder for users to narrow down their results.
Design Challenge
The search results page is more than a product list. It's also an important entry point connecting users' shopping needs with business opportunities. The challenge was to balance product discovery, shopping criteria, featured services, and commercial exposure within limited screen space.
03
GOAL
Help people find the right product faster
01
Bring results closer to what people are actually searching for.
02
Help users explore related products and narrow down results with more relevant search suggestions and filters.
04
STRATEGY
Different search intents need different ways to explore
People don't always know what they want to buy at the start. Searching "matcha" may still be about browsing product types; searching "water bottle" often means comparing specifics like capacity and material.
So I proposed two design strategies to support different search needs.
Strategy 01
When intent is still forming
Intuitive images and category entry points help people recognize product types quickly and find their next direction to explore.
Strategy 02
When intent is specific
Combining the search keyword with product attributes, filters adapt to the search context and surface the information that matters for the decision.
05
DESIGN
Key design decisions
01
Based on user research and search data, we found that users often searched with broad keywords, making it difficult to narrow down their results.
To address this, I designed visual search suggestions that combine images and text, helping users explore relevant product categories and find a clearer direction.
For example, when searching for gifts, users could quickly explore categories like birthday gifts, gifts for women, or Christmas gifts without repeatedly changing their search keywords.

Supporting Search Guidance Beyond the Top of the Page
I also introduced related search suggestions within the product grid and at the bottom of the page, extending search guidance beyond the initial entry point.
By integrating these suggestions into the browsing experience, users could refine their searches and discover more relevant products without starting over.

Related Searches in the Product Grid
Integrated related search suggestions into the product grid, allowing users to discover alternative search directions while browsing.

Related Search Keywords at the Bottom
Added related keywords at the bottom of the page, giving users another opportunity to refine their searches and continue exploring.
02
Different products require different filters, but the previous experience showed the same filtering options for every search.
To address this, I redesigned the filtering experience across devices, incorporating dynamic filters based on search keywords and product attributes.
On mobile, I designed a two-row layout. The first row displayed basic filters, while the second showed relevant product attributes based on search keywords.
On desktop, dynamic filters were integrated into the left sidebar, allowing users to refine their results while browsing.
I also collaborated with the PM, AI, Localization, and Business teams to review the recommended attributes and ensure they aligned with real shopping needs.

03
One of the business requirements was to create more opportunities for brand advertising on the search results page.
I noticed unused space beside the product grid on desktop, so I proposed placing ads there without taking up the main product browsing area.
This approach created additional advertising opportunities while maintaining a balance between product discovery and business goals.
06
OUTCOME
PRODUCT IMPACT
01
02
03
Visual Search Refinement appeared in approximately 15–20% of searches, achieving a 12% click-through rate on Desktop Web, nearly twice that of other platforms. This suggests stronger engagement with search guidance on Desktop Web.
Across all four platforms, users who clicked visual search suggestions had product view rates 8–16 percentage points higher than those who did not. These findings suggest that visual search guidance goes beyond offering additional search options, with the potential to help users narrow down their search and continue exploring products.
07
TAKEAWAYS
Rethinking how search results help people choose
01
Even when users know what they want to buy, they may not know which keywords to use. This project reinforced my belief that a good search experience should not only provide relevant results but also help users find the right direction.
02
Based on these findings, we could explore two improvements:
• Expand feature coverage: Make dynamic filters available for more search terms.
• Improve filter visibility: Make the filter entry point easier to discover and use.

