Designing a unified filtering system
for faster hotel and ticket search
for faster hotel and ticket search
Ozon Travel B2C
Client
Stakeholder & User interviews,
low to high-fidelity wireframes, prototyping, UX processes
low to high-fidelity wireframes, prototyping, UX processes
Methodologies
January, 2025
Date
Overview
This project focused on improving the speed and efficiency of choosing hotels and flight tickets through an updated filtering system. The goal was to create a unified, scalable solution that simplifies discovery, reduces time-to-selection, and supports both short-term updates and long-term product growth across travel verticals.
Problem
The existing filtering systems for hotels and flights were inconsistent, overloaded with options, and not optimized for user priorities.
Users often struggled to find relevant results quickly due to non-intuitive grouping, redundant filters, and differences between verticals.
From a technical side, fragmented logic complicated development, slowed down feature releases, and hindered scaling.
These issues led to longer search times, decreased filter usage, and lower conversion rates.
These issues led to longer search times, decreased filter usage, and lower conversion rates.
Process
- Defined a unified filtering logic covering shared and unique attributes for each vertical.
- Created and validated reusable UI patterns aligned with design system principles.
- Proposed a phased rollout strategy. Phase 1: Update existing filters using new structure and hierarchy. Phase 2: Gradually introduce new filters when ready on the backend.
- Prepared interactive prototypes and tested flows with real users to optimize hierarchy and placement.
- Designed a unified component structure to speed up implementation and ensure consistent behavior across verticals.
- Partnered with analytics and product teams to run A/B tests for each new filter release and track performance impact.
Research
To clarify user needs and business priorities, I conducted comprehensive research that included:
- Current-state analysis: Reviewed how filters currently work in both hotel and flight flows.
- Competitor benchmarking: Studied leading travel apps and booking platforms to identify common filtering patterns.
- Usage analytics: Collected data on most frequently used filters, retention within the filtering panel, and drop-offs.
- Pain points mapping: Identified overloaded or irrelevant filters that did not contribute to conversion.
Goals
— Improve search efficiency and conversion through a more intuitive filter system.
— Create a unified design and logic for both the hotel and flight verticals.
— Build a foundation for scalability (reuse in rail and tour verticals later).
— Increase adoption and usage of filters via A/B-tested launches.
— Reduce design and development complexity through standardized components.
— Create a unified design and logic for both the hotel and flight verticals.
— Build a foundation for scalability (reuse in rail and tour verticals later).
— Increase adoption and usage of filters via A/B-tested launches.
— Reduce design and development complexity through standardized components.
Result
Benefits
Conclusion
By combining research-driven insights, iterative design, and system thinking, I built a robust and flexible filtering framework that improved both user experience and business performance. The new system not only enhanced current usability and conversion metrics but also established the groundwork for future growth and operational efficiency.
— Unified filtering experience across multiple verticals.
— Improved discoverability and relevance of search results.
— Reduced cognitive load and friction for users.
— Scalable foundation supporting future expansion to rail and tour products.
— Accelerated development thanks to standardized, documented components.
— Improved discoverability and relevance of search results.
— Reduced cognitive load and friction for users.
— Scalable foundation supporting future expansion to rail and tour products.
— Accelerated development thanks to standardized, documented components.
A consistent, user-friendly filtering system was successfully launched for both hotel and flight search. Key improvements:
+35% increase in filter usage rate.
–12% reduction in average search time per session.
Increase in conversion to purchase after applying filters.
2× faster implementation time for new filter types due to reusable components.
+35% increase in filter usage rate.
–12% reduction in average search time per session.
Increase in conversion to purchase after applying filters.
2× faster implementation time for new filter types due to reusable components.
