Revamping Search and IA
How Redesigning an App's Information Architecture Transformed the Trail-Finding Experience
My role:
UX/UI Designer
Company:
SpringTech
Bivy
Timeline:
+ 6 months
Team:
Product Manager
UX/UI Designer
Developer
The Struggle
Bivy's Great Selection of Adventures Fails to Impress
The Bivy app catalogues 50,000 trail adventures.. However, with just 1,000 monthly visitors and a high bounce rate, it was clear that adventurers were getting lost, not in the wilderness, but in the app's interface. The challenge was not just to attract users but to guide them effortlessly through their journey of discovery adventure trails.
Users do not find the search/filtering experience helpful
Users are not browsing and discovering new adventures
Shifting the Focus
Why was the Bounce Rate so high?
Users left the Bivy App, but I didn’t know why. After a cognitive walkthrough with 3 users, I found the following:
Simplified navigation is crucial for effective site exploration
Weather filters and trail advisories are essential.
User-contributed reviews and photos guide expectations.
Community updates directly from users are valuable.
Specific filters to improve search accuracy.
Stakeholder Perspective:
Clients desire a modern aesthetic with functional filters. But through cognitive walkthroughs, structural flaws were exposed:
Unclear organization and labelling hinder navigation.
Inconsistent navigation paths lead to dead ends
Grouping content is essential for relevance and discovery.
Filters need clarity and efficiency.
Space utilization must be optimized to avoid clutter.
Convincing the client to expand the scope of the project
Aware of the client's passion for hiking, I leveraged a map analogy to convey the necessity for a website overhaul. The client was receptive to undertaking a larger overhaul.
Looking for the solution
Steps for Improving Information Architecture
I planned out the project into several actionable steps:
1. For enhancing the adventure browsing experience:
Simplified navigation is crucial for effective site exploration
Weather filters and trail advisories are essential.
User-contributed reviews and photos guide expectations.
Community updates directly from users are valuable.
Specific filters to improve search accuracy.
2. For optimizing the search for a specific adventure:
Organize autocomplete suggestions into categories for quicker access.
Present relevant and dynamic filters prominently above search outcomes.
Allow users to create and apply their custom search presets.
Emphasize search terms within results to facilitate rapid review.
Enhance keyboard navigation with efficient shortcuts and auto-focus on the first result.
Integrate interactive maps for easy adventure or geography selection.
Expand filters for tailored searches, enhancing user relevance and personalization.
Transforming IA
Improving the Browsing Experience through Card Sorting and Tree Testing
Leveraging user insights from card sorting and tree testing, I refined the platform's architecture, resolving navigational issues and enhancing the browsing journey
Card Sorting
Using OptimalSort, I explored user interactions and feedback to uncover confusion areas. This analysis led to actionable steps for improvement:
Enhance the blog for community and learning
Incorporate real-time features
Establish a dedicated section for educational and safety content
Revise of the content on the home page
Tree Testing
Tree testing shed light on the effectiveness of the proposed information architecture by testing how well tasks could be completed with the current sitemap. Two significant areas for improvement were identified:
Difficulty in locating the company's return policy led to integrating a direct link in the product description section, thus streamlining access and improving navigational clarity.
Despite users placing the search adventures card in the adventures section, the process was too slow. To remedy this, a search field was added to the home page for quicker access and enhanced browsing efficiency.
Refining Search for Adventure
A Comprehensive Strategy for Personalized Exploration
In refining the search suggestions, I've integrated a comprehensive approach considering relevance, popularity, user activity & preferences, geographical closeness, seasonal context, ongoing promotions, variety, specific filters, current availability, and the freshness of adventures. This ensures that:
Matches align tightly with search terms, favoring exact beginnings.
More engaged and popular adventures rank higher.
Personalization reflects users' historical interactions and choices.
Nearby adventures are prioritized for convenience.
Results adapt to the current season and weather.
Promotions and events temporarily boost certain adventures.
A diverse array of adventure types caters to broad interests.
Selected filters refine the search focus.
Only accessible and available options are shown.
Fresh content gets a spotlight for up-to-date offerings.
The Path to Enhanced Usability
Choosing the Right Design for Adventure Seekers
To identify the best design for an improved user experience, I conducted a test comparing two design variations. Participants, selected for their passion for outdoor activities, were tasked with finding an adventure that met specified criteria, providing a practical assessment of each design's effectiveness.
Key findings:
Most found the correct adventure within 2-3 minutes.
Success rates: Design 1 at 62.5%, Design 2 at 87.5%.
SUS scores: 67.5 for Design 1, 82.5 for Design 2.
Design 1 described as dark, impersonal; Design 2 as welcoming, clear.
Results adapt to the current season and weather.
Based on superior user feedback and success rates, Design 2 was the clear choice.
How I'd Change Things Up: A Reflective Approach
After wrapping up this project, I've picked up a lot of new insights. If I were to dive into it again today, here's how I'd tweak my design process to make it even better:
Analyzing Search Logs: I would review search logs to identify common search queries, frequent no-result searches, and patterns that indicate user struggle. This data can help refine the search algorithm and improve result relevance.
Search Quality Evaluation: I'll analyze the top 100 most-searched queries, manually assessing the relevance of the leading results for each. By scoring their relevance and aggregating these scores, we can measure search result quality. High scores signify excellent search performance, while lower ones highlight improvement needs. Tracking these scores over time or against industry benchmarks helps monitor progress or understand our competitive standing.
