Viu Search Experience
Viu offers a world-class catalogue of premium Asian movies, tv shows and anime, but this does not matter if users cannot find them. The design team discovered that the search journey was not optimized to help users reach their programs, so we set out to fix that.
Summary
In my work to improve Viu's search experience, I conducted user research and competitive analysis to identify key challenges. I focused on enhancing how content is tagged and organized within the platform (known as CMS tagging) to make it easier for users to find what they're looking for. Additionally, I provided visual search guidance, created a more informative no-results page, and implemented multilingual search results to help a diverse audience discover and enjoy the content they love.
My Role
-
Conduct quantitative and qualitative research, including user interviews, to understand user behaviour and industry standards
-
Create user flows, including edge cases, for all scenarios
-
Create high-fidelity wireframes and prototype
-
Collaborate with product managers and developers to understand technical feasibility and constraints and ensure business and user needs are met
Design Process
-
Design brief review with design team and product manager
-
User research, including comparative and competitive analysis, keyword analysis, user interviews and Hotjar tracking
-
Design workshop- User empathy mapping, How Might We's, Crazy 8's
-
First draft mockups
-
Usability testing with design team and ops
-
Iterate on designs
-
Final review (head of engineering, product and operations teams)
-
UAT and Visual QA
Problem
Our users cannot find the shows they're looking for via search bar and are giving up
Goal
To help our users achieve their goals and find the programs they love
Users
Our active user base of viewers, reaching for programs to watch
How Might We?
....help users reach their target results?
...make users feel clear about what they are searching?
...help users find the most relevant results in the easiest way?
...keep users from getting stuck on the no-results page?
Understanding our Users
I conducted qualitative and quantitative research to better understand our users' search behaviors and goals, and to understand how Viu's competitors and other top companies handle their search function.
1. Users' Search Queries Analysis
Summary: I performed keyword analysis by examining the search terms entered by users to improve content discoverability on the platform.
Elaboration: By carefully reviewing the search terms users frequently entered, I identified common patterns and gaps in the existing keyword tagging. This analysis allowed me to refine and optimize how content was tagged, ensuring that search results more accurately reflected user intent. For instance, I discovered that users often entered incomplete titles and phrases, indicating a need for a search system that could handle such variations more effectively. Additionally, users frequently searched for program titles over other categories, highlighting the importance of making this content more accessible.
In response to these insights, I updated the keyword tagging strategy to better accommodate partial matches and prioritized program titles in search results by implementing an "All" results page. These adjustments led to a more intuitive search experience, as users found relevant content more easily and experienced fewer frustrations with incomplete searches. Overall, these improvements contributed to a more seamless and satisfying user experience, aligning the search functionality more closely with user needs and expectations.
2. Competitive Analysis
Summary: I conducted a competitive analysis of search experiences on platforms like Netflix, Disney, Apple, Spotify, HK Movie 6, and YouTube to understand how they manage large volumes of data and optimize search functionalities.
Elaboration: By examining how leading platforms handle search and data, I identified effective strategies for organizing and conducting searches. Notably, these platforms use advanced filtering and recommendation systems to enhance user satisfaction. For instance, Netflix and Spotify leverage personalized recommendations based on user behavior, while YouTube combines search and recommendation algorithms to effectively surface content.
These insights reveal that adopting industry-standard practices, such as intuitive search interfaces and sophisticated data handling, can significantly improve our own search functionality. Instead of creating entirely new solutions, we can integrate these proven strategies into our platform, ensuring a more efficient and user-friendly search experience.
3. Unbiased User Screen Recordings on Hotjar
Summary: Hotjar session recordings, screen recordings of unbiased users interacting with Viu, revealed that users frequently bypass the search bar in favor of using preset categories or accessing content through external sites like Google.
Elaboration: This user behavior underscored several key areas for improving our search functionality. To address these issues, we implemented several targeted enhancements:
-
Keyword Tagging and CMS Improvements: We refined how content is tagged and organized within the database to ensure more accurate search results that align with user intent.
-
Content Prioritization and Display: We adjusted the way content is prioritized and displayed on the search results page, making it easier for users to find relevant content quickly.
-
Multi-Lingual Support: We enhanced our search capabilities to support multi-lingual queries. Unfortunately, Viu's language sites were characterized by separate databases, making it difficult to search and display multilingual results at once. Due to constraints with budget and time, combining these databases was not feasible. Instead, we created a feature that would allow the system to detect another language and offer a path to searching in the other language database, catering to a diverse user base and ensuring that language barriers do not impede content discovery.
4. User Interviews
Summary: User interviews revealed that users prefer to search using incomplete titles and phrases to save time, and Chinese users often conducted searches in multiple languages. This behavior aligns with findings from keyword analysis, which also showed a prevalence of incomplete search terms. Additionally, Viu’s search function did not support multilingual searches.
Elaboration: Insights from user interviews and keyword analysis highlighted critical areas for improving the search experience on Viu:
-
Handling Incomplete Titles and Phrases: Both the keyword analysis and user interviews indicated that users frequently enter incomplete titles and phrases. This behavior is driven by a desire to save time and avoid hassle. In response, we suggested to enhance the search functionality to better handle partial queries. This includes implementing fuzzy matching and autocomplete features to recognize and process incomplete search terms more effectively, ensuring users can quickly find relevant content.
-
Multilingual Search Capability: User interviews revealed that Chinese users, in particular, often searched using a mix of English and Chinese, or simply in English on the Chinese site because it was simpler. To address this, we upgraded our search system to support multilingual queries, incorporating language detection and indexing to ensure that users can search seamlessly in multiple languages.
These improvements would address the combined insights from user interviews and keyword analysis, leading to a more efficient, user-friendly, and inclusive search experience on Viu.
4. Empathy Mapping
Summary: Using insights from user interviews, my team and I conducted a workshop to map users’ responses based on what they said, did, and felt. This exercise helped us identify two critical areas for focus: searching and discovery.
Elaboration: During the workshop, we organized user feedback into categories based on what users said, did, and felt. This analysis revealed significant challenges in searching for specific content and discovering new content. In response, we focused on improving search functionality by enhancing content organization, tagging, and multilingual capabilities. We also refined the discovery process through various content suggestion methods—such as hot search, trending search, and search history—and optimized how search results are organized and displayed, making both processes more intuitive and user-friendly.
Challenges
Working on this project came with its share of challenges, particularly in terms of team dynamics and technical limitations.
One of the primary challenges was collaborating with a team manager who frequently pushed ideas without a research or UX background. His suggestions were often based solely on personal opinions, without any grounding in user data or design principles. This created friction within the team, as we wanted to ensure that our design decisions were informed by solid research and aligned with best practices.
To navigate this, we had to find a balance. While he had the final say, we incorporated his ideas but adapted them in different styles that would make more sense from a UX perspective. Additionally, we made it a point to present our research and findings whenever possible, which helped persuade him to see the value in our approach and sometimes adjust his stance.
Another significant challenge was the technical limitations of our development team. The team's capabilities were not advanced enough to combine different CMS databases, which meant that the different language sites had to remain completely separate. This presented a hurdle in handling multi-language search queries effectively. To address this, we implemented a workaround by adding a button that allowed users to search across different language databases, ensuring a more seamless experience despite the technical constraints.
Solutions
1. Proposing Enhanced Tagging and Optimized Content Prioritization
Rationale: User feedback and keyword analysis revealed that search results often didn't align with user intent due to inconsistent tagging and poor content prioritization.
Solution: I proposed a refined tagging strategy and designed an intuitive content layout that categorizes content effectively and supports multilingual searches. I introduced a feature where the system detects foreign languages and offers a chip/button to search the corresponding language database.
Impact: These combined efforts ensured that users could quickly and accurately find the content they were looking for, improving search efficiency and satisfaction.
Previous Results Page
2. Developing a User-Friendly, Visually Guided Search Interface
Rationale: Users frequently struggled with search queries that were incomplete or misspelled, leading to poor results and frustration.
Solution: I designed a friendly search interface that offers visual guidance while users type. As they enter a query, relevant options appear below, with the text they've already typed highlighted in a different color. This helps users see their progress and refine their search before hitting enter.
Impact: This design not only improved search accuracy but also enhanced the user experience by making the process more intuitive and reducing errors.
3. Enhancing Content Discovery and Guiding Users on No-Results Pages
Rationale: To prevent users from getting stuck when no search results were found, it was important to offer alternative content and discovery options.
Solution: I designed features that guide users through no-results pages by offering personalized content suggestions, trending searches, and related topics. This ensures that users can continue their journey without hitting a dead end.
Impact: These features increased user engagement by keeping users on the platform longer and helping them discover content they might not have initially searched for.
Previous No Results Page
