Rakuten Beauty
AI-Driven Features for Rakuten Beauty
Evolving a salon directory into a more personal, intent-aware discovery assistant.
- Role
- UI/UX Designer
- Team
- Rakuten Beauty Design Team
- Responsibilities
- Product concept, search experience design, AI interaction design
- Platform
- Web
- Tools
- Figma
The opportunity
Nuanced needs do not fit neatly into checkboxes.
Rakuten Beauty’s discovery model relies heavily on rigid filters and exact keyword matching. Users with detailed needs must cross-reference filters, scan long result lists, and read many individual reviews to understand whether a salon is right for them.
I explored how generative AI could move the service from a search tool toward a personalized discovery assistant, shortening the path from intent to booking while maintaining familiar booking patterns.
Feature 01
Semantic search translates intent into useful matches.
Instead of checking a series of boxes, users describe their ideal experience in natural language. The system identifies requirements such as location, technique, and atmosphere, then returns focused results.
Each result includes a match score and a short explanation, helping users understand why the salon fits their request and building trust in the recommendation.
Feature 02
Recommendations make discovery proactive.
A personalized feed uses past bookings and search behavior to surface salons the user may not have found. Match scores shift the experience from actively searching to receiving a curated set of relevant options.
Feature 03
Review summaries reduce choice paralysis.
Instead of checking a series of boxes, users describe their ideal experience in natural language. The system identifies requirements such as location, technique, and atmosphere, then returns focused results.
Each result includes a match score and a short explanation, helping users understand why the salon fits their request and building trust in the recommendation.
Visual language
AI that feels alive, but still feels like Rakuten Beauty.
Fluid pink and blue gradients extend the existing brand palette and signal where AI is actively interpreting or synthesizing information. High-contrast type and clear calls to action keep the functional layer accessible.
Impact
A seamless evolution of the service, powered by AI
While these features are currently in the conceptual phase, the design framework provides a clear roadmap for how Rakuten Beauty can transition from a traditional directory into an AI-first discovery engine. By implementing this vision, the platform could expect to see the following:
- Lowered Barrier to Entry: Replacing complex filtering with a single search bar reduces the cognitive load for new users, likely increasing the conversion rate from search to booking.
- Proactive Engagement: Moving from reactive search to proactive Match Scores for salons turns the platform into a personalized assistant, deepening user loyalty.
- Enhanced Decision-Making: By synthesizing hundreds of reviews into a single summary, choice paralysis can be removed, helping users feel confident in their selection much faster.
AI Powered Search Prototype
To demonstrate how the AI Powered Search would look like, I have created a prototype of the feature below. Clicking on the search field and then the search button will simulate a user entering a search query and running the AI search.