YouTube AI Recaps

YouTube AI Recaps is a smart summarization feature that helps users catch up on long-form videos by generating timestamped key takeaways in seconds.

Role

UI UX Designer

TimeLine

April 2025 – May 2025

Tools

Figma, FigJam, Notion

Project Type

Bootcamp project, UX Academy

The Problem

Users often struggle to watch long videos start to finish because:

• 🧱 Traditional search tools feel clunky and inefficient

• ⏳ Users waste time skipping through to find relevant moments

• 😵‍💫 Lack of structure in long videos makes retention difficult

• 🤯 Passive watching leads to low engagement and knowledge loss

The Solution

YouTube AI Recaps simplifies video engagement by:

• 🎯 Summarizing videos with one clean, AI-generated recap

• ⚡ Giving users timestamped takeaways for quick browsing

• 🔁 Helping users review content without rewatching entirely

• 🧠 Supporting retention through structured, digestible info

The Research

For the YouTube AI Feature, I conducted targeted user research to understand how people interact with long-form video content and what prevents them from getting value quickly. This included a light competitor review of tools like Glasp and Eightify. I also conducted user interviews to uncover common pain points around content skimming, retention, and daily YouTube habits. From these insights, I created a focused user persona to shape our design direction. I then translated our findings into an affinity map to group insights by motivation, frustration, and behavior patterns — ensuring every design choice solved a real user need.

To learn how users consume and retain video content, I interviewed 8 participants. Key findings:

  • 90% said they skip around videos instead of watching fully

  • Most users rely on guesswork when looking for useful moments

  • Many users said they forget what they watched shortly after

User Interviews 
Competitive Analysis

I analyzed tools like Eightify, Glasp, and YouTube’s own transcript feature to identify gaps. Findings:

  • Eightify offers fast AI summaries directly on YouTube

  • Glasp enables highlight saving and some topic filtering

  • YouTube’s built-in transcript is native and easily accessible

Strengths
Weaknesses

  • Summaries often lack context or nuance from the video

  • Most interfaces feel text-heavy and unintuitive

  • No clear path to organize or review important takeaways

Eightify

  • Instantly generates AI-powered video summaries on YouTube

  • Provides time-stamped bullet points

  • Easy to use via a browser extension

Glasp

  • Lets users highlight, save, and tag video notes

  • Topic filtering and personalized highlight feeds

  • Integrates with other learning tools (Notion, Readwise)

YouTube Transcript

  • Automatically generates transcripts for most videos

  • Allows word-level search and navigation

  • Native and always available within the YouTube UI

Core Features

After analyzing competitors and gathering insights from interviews, I created two user personas to represent the primary mindsets and behaviors uncovered during research.

User Personas

Age: 35
Occupation: Project Manager
Location: Detroit, MI
Family: Married No kids

Marcus Green

Goals:

Stay sharp by consuming high-quality, relevant information

Maximize productivity without burning out

Advance personal and professional growth

Pain Points:

Gets frustrated when content feels bloated or overly padded

Feels like digital tools often try to do too much

Doesn’t have patience for features that require too much onboarding

Age: 20
Occupation: Full-time college student
Location: PHX, AZ
Family: Lives on campus with two roommates

Talia Nguyen

Goals:

Discover new ideas, content, or tools that spark personal growth

Make the most of her limited time between responsibilities

Stay informed without feeling behind in classes or news

Pain Points:

Struggles to focus when information is presented slowly

Feels overwhelmed by rigid or multi-step routines

Finds it difficult to stick to long-term morning habits

To make sense of the interview data, I created an affinity map — a method for synthesizing user quotes by grouping related thoughts, pain points, and behaviors. This allowed me to organize raw feedback into clear, actionable patterns.

Affinity Map

I organized the observations into clusters like benefits, common findings, pain points, and possible solutions. This helped highlight what mattered most to users and revealed key opportunities for the product.

Insights

Just One is grounded in key research insights, including:

  • Users want a lightweight and flexible routine that doesn’t feel overwhelming

  • Positive reinforcement and a sense of progress are key motivators

  • Switching between multiple apps (alarms, journals, lists) creates friction

  • A consistent morning routine helps users feel in control and focused

After completing the affinity map, I moved beyond research and into the next phase — Prioritization & Roadmapping — where I translated insights into actionable design decisions.

Project goal & Roadmapping

To guide the direction of this feature, I started by clearly defining the project goals and outlining a realistic task flow. The goal was to give users instant clarity on whether a video was worth their time. From selecting a YouTube link to generating a concise, timestamped summary, each step in the flow was designed to feel effortless and focused. This foundation helped me prioritize what mattered most and ensure the experience was both functional and intuitive.

To guide the design of this feature, I established a set of clear project goals focused on clarity, speed, and user control:

  • Allow users to quickly decide if a video is worth watching

  • Give control over summary length with an intuitive slider

  • Display clear, timestamped bullet summaries

  • Keep the experience fast, focused, and clutter-free

  • Help users retain and revisit key moments from the video

Project Goals

After defining these project goals, I mapped out a task flow to visualize how users would move through the YouTube AI Summarizer experience — from landing on the feature to receiving their final summary. This helped ensure every interaction supported the core goals while maintaining simplicity and speed.

Task Flows

Summarize video

After mapping out the task flow, I had a clear understanding of how users would move through the feature. This allowed me to begin shaping the wireframes with intention—focusing on simplicity, hierarchy, and clarity. I wanted every screen to feel purposeful, ensuring that key actions like adjusting the summary length or accessing timestamps were both intuitive and visually accessible.

Wireframes

With the core experience defined and the task flow locked in, I shifted my focus to translating structure into layout. I began by sketching low-fidelity wireframes to quickly explore how users would move through the feature. These initial drafts helped me clarify hierarchy, prioritize essential actions, and ensure the layout supported intuitive interaction. Once the flow felt right, I translated these sketches into high-fidelity wireframes to visualize the polished experience and guide the next phase of testing.

Before diving into high-fidelity design, I reflected on how the interface should feel—focused, easy to use, and grounded in simplicity. I wanted the wireframes to visually reinforce the tool’s purpose: making AI summaries accessible and clear. This meant using minimal distractions, intuitive layout patterns, and clear touchpoints that aligned with user expectations and feedback.

Lofi Sketches

After creating low-fidelity wireframes to explore layout ideas and user interaction, I moved straight into high-fidelity wireframes—refining spacing, typography, and visual hierarchy to better simulate the final product and prepare for usability testing.

Hifi Wireframes

These high-fidelity wireframes brought the feature to life—uniting functionality and flow into a clean, focused experience. Every detail reinforced the goal of making summaries feel accessible, clear, and time-saving. The interaction model now felt streamlined and purposeful, aligning well with users' needs for efficiency and clarity.

With the experience fully structured, it was time to test how real users interacted with the summarizer. I used these sessions to fine-tune layout, catch minor friction points, and ensure the feature felt seamless and helpful in actual usage—just as it did in planning.

Iterations and Testing:

Once the high-fidelity designs were complete, I moved into usability testing by interviewing and observing real users as they interacted with the prototype. The goal was to uncover friction points, see what felt intuitive, and hear honest reactions to the flow and features. I combined task-based testing with quick interviews to get both behavioral insight and verbal feedback. These sessions revealed what was working and what still needed refinement — helping me prioritize meaningful improvements backed by real usage.

To clearly capture insights from these sessions, I created a scorecard to track task success, timing, and direct user comments.

Usability Test Scorecard

From this testing, it became clear that while users understood the feature’s purpose, there were key opportunities to improve usability and interaction clarity. I focused my iterations on enhancing feedback and accessibility: adding a visible “Set Summary Length” label to eliminate confusion, and introducing a speaker icon to let users listen to summaries aloud. These updates made the feature feel more intuitive, inclusive, and polished—bringing the experience closer to real-world usage needs.

Iterations

1. Summary Interface Enhancements

Users were unclear on how the summary length slider functioned and missed that audio playback was available. To improve clarity and engagement, I added a label above the slider to explain its purpose and introduced a speaker icon that lets users listen to the summary out loud — making the experience more accessible and intuitive.

Conclusion

Designing the YouTube AI Summarizer wasn’t just about speeding up video consumption — it was about giving users back control of their time. By focusing on clarity, precision, and personalization, I crafted an experience that transforms long-form content into digestible, high-impact insights. Every detail — from layout to interactions — was shaped by real user needs and tested with intention. The result is a seamless feature that empowers users to get the value of a 15-minute video in seconds, without losing context.

Next steps

  • Explore more granular summary controls, such as topic segmentation or chapter tagging

  • Add a “Save Summary” feature for quick reference or note-taking

  • Introduce AI personalization so summaries reflect individual preferences or history

  • Continue testing with diverse user groups to refine accessibility and mobile usability

Key Takeaways

  • Users value speed and clarity — summarization must feel instant and frictionless

  • Control matters — sliders, toggles, and customization help users feel in charge of their experience

  • Summaries must maintain context — users want condensed content, but not at the cost of meaning

  • AI is only as good as its UX — clean design and smart defaults turn complex tech into an intuitive tool