YouTube Homepage Swipe Feature
Overview
Role: UX/UI Designer
YouTube provides an immersive entertainment experience with a vast library spanning genres like music, gaming, science, and more. It offers features such as autoplay, personalized recommendations, and playlists that help keep users engaged. However, while YouTube's algorithm delivers content based on user behavior, it doesn't always reflect what users want to see in the moment, beyond liking and subscribing.
Although YouTube's algorithm is designed to personalize content delivery, users can feel disconnected from what’s being recommended. Often, they find it challenging to immediately influence the content appearing on their homepage beyond liking, subscribing, or commenting. This leads to frustration when users want to see specific types of content right away, but the algorithm doesn’t adjust quickly enough.
The goal of this project was to address this pain point by designing a new feature that enables users to guide the algorithm with greater accuracy and immediacy. This would not only improve user satisfaction but also help retain users by keeping them engaged with the content they want to watch.
Research
I analyzed competing platforms, such as Instagram’s “Explore” feature and TikTok’s recommendation system, to see how they allowed users to influence content delivery. By understanding their strengths and weaknesses, I identified opportunities for YouTube to implement a more user-guided experience, without overwhelming users with options.
I conducted interviews with three users who frequently engage with YouTube’s platform or any of the competing social media platforms. The goal was to understand their behavior, preferences, and pain points while navigating new content. Key insights revealed that users were often finding themselves doom scrolling or refreshing the page until they would find content they would be interested in watching, along with rewatching content to adjust their feed. One of the three even prefers another social media platform over YouTube’s due to the ease of discovering new content that pertains to their interests in the moment.
After gathering feedback, I used FigJam to create an affinity map that helped me group common themes from the interviews. Key takeaways included users’ desire to skip or see less of a particular content, the ability to quickly change the topic or genre of new discoverable content at any given moment, as well as finding desirable content quickly.
Based on the affinity mapping, I developed Point of View (POV) statements and How Might We (HMW) questions to frame design opportunities.
“How might we empower users to fine-tune and adjust their recommended content effortlessly, ensuring they can control their video platform experience based on their evolving interests?”
“How might we give users more control to filter out irrelevant or uninteresting videos, allowing them to shape their content recommendations and eliminate misleading or unrelated content effectively at the homepage?”
“How might we design a flexible content stream feature that aligns with what the user would like to see at the moment, delivering content exclusively tailored to their interests?”
Persona
From the research, I developed a persona representing an avid viewer who uses YouTube for both work and relaxation. This persona helped guide design decisions by keeping the users' goals and frustrations at the forefront.
Using the personas and data gathered, I designed a user flow to illustrate how users could interact with the new feature. The flow needed to be simple and intuitive, allowing users to adjust their content preferences without leaving the homepage or interrupting their current viewing experience.
Wireframes
I created low-fidelity wireframes in Figma to explore different ways to incorporate this feature. After feedback from an initial concept review, I refined the wireframes to reflect a simple swipe gesture on the homepage that gives the user more control of the algorithm.
Following the low-fidelity wireframes, high-fidelity prototypes were developed, focusing on key interactions:
Swipe for Interest: Users could swipe right to indicate interest, refining recommendations in real time.
Not Interested & Additional Options: Swiping left dismissed content or provided options like blocking channels for better content curation.
Undo Action: Allowed users to revert accidental swipes, ensuring greater control over content selection.
Preference Management: Enabled users to remove previously disliked videos, fine-tuning their recommendations.
These refinements enhanced user control and personalization, ensuring a seamless and intuitive experience before usability testing.
Usability Testing
User testing involved three participants who were given three tasks to complete using the prototype:
Swipe to mark a video as “Interested.”
Swipe to mark a video as “Not Interested” or access additional options.
Remove a previously disliked video from preferences.
The usability test results provided key insights into the effectiveness of the design, highlighting both successful elements and areas for improvement.
Usability Test Results
Success Metrics:
Task Completion: All participants (3/3) successfully completed all tasks.
Task Duration: All participants (3/3) completed each task in under 2 minutes.
Ease of Use: All participants (3/3) found the feature simple, self-explanatory, and easy to understand.
Frequency of Errors:
All participants (3/3) struggled to understand how to initiate the first task and required hints or prompts.
All participants (3/3) encountered no issues with the second task.
Two participants (2/3) suggested consolidating swipe actions into one direction instead of separating them.
One participant (1/3) noted that selecting “Interested” caused the video to disappear, which was unintended.
All participants (3/3) experienced difficulties completing the third task due to a design flaw in the task flow.
User Satisfaction:
Task #1 Average Rating: 6.6/10
Task #2 Average Rating: 9/10
Task #3 Average Rating: 7.3/10
These results revealed key areas for refinement, including improving the clarity of user actions, enhancing the task flow, and adding visual feedback.
Next Steps
Improve onboarding and guidance:
Implement an overlay or pop-up message introducing the new feature.
Refine swipe interaction:
Explore consolidating swipe actions into a single direction.
Prevent videos from disappearing immediately after marking them as “Interested.”
Enhance task flow for removing disliked videos:
Adjust the design to make the process more intuitive.
Increase user feedback visibility:
Add confirmation pop-ups or visual indicators to confirm actions.
Expand customization options:
Introduce a comment box in the “Tell Us Why” section for users to provide additional feedback.
These findings reinforced the importance of clear user guidance and intuitive design. While the feature successfully improved content control, further refinements would enhance usability and engagement. Moving forward, additional testing and iterations would help ensure a seamless user experience before full implementation.
Iteration
The prototype successfully achieved its intended goals, even though it was not developed or beta-tested within the live app. While implementation may present challenges on the algorithmic and technical side, I am proud of the feature’s simplicity and its potential to enhance user engagement. This project pushed me to think creatively within the YouTube/Google brand identity, apply advanced design methods, and refine my problem-solving approach. I would love to see this feature implemented, gather real user insights on its effectiveness, and observe how it influences YouTube’s recommendation system over time.