Throughout my career and most recently at YouTube, I’ve learned that there’s no single definition of innovation. Instead, there are many different yet equally valid ways to introduce new ideas, regardless of your specific process or mindset. What we’ve discovered at YouTube—over many years and many projects—is that our teams, products, and ultimately our users, greatly benefit when we use multiple, complementary forms of innovation.
Our product teams focus on two approaches: versioning and visioning. Versioning uses incremental product improvements that add up over time to significantly improve user experience. Visioning, in contrast, is an approach that addresses more open-ended questions and emphasizes longer-term thinking. At YouTube, we use visioning to explore uncharted terrain in the form of new users and/or new problem spaces. While perhaps lesser known, visioning is a technique you too can use to speed up product innovation.
In 2014, we established a dedicated visioning team with a mission to help the company and our product teams explore new terrain and make bigger leaps a reality. This team—which I lead—focuses on exploring new areas, defining new features or products, and rethinking our existing products. Over the years our team has helped lay the groundwork for products such as YouTube Kids and YouTube Go, and launched new features like our community tab and interactive cards. Let’s take a closer look at visioning in practice.
A vision for learning on YouTube
Every day, millions of people come to YouTube to learn something new, from math, science, and literature, to language lessons, music tutorials, and test prep. In early 2018, YouTube created a newly dedicated product and engineering team focused on these forms of learning, and partnered with the visioning team to develop a long-term vision and product roadmap for a newly imagined learning environment. To achieve this, we used a 6-phase process that we’d developed and refined over many projects. Now we’ll dig into each step of the process from building a team of collaborators and getting to know the problem, to launching an entirely new product into the world.
The visioning journey can be broken down into six discrete phases: 1. Activate the team; 2. Immerse yourself in the problem; 3. Frame the opportunity; 4: Explore the possibilities; 5: Define the vision; 6: Inspire action.
Phase 1: Activate the team
Before we even embarked on the visioning journey, we first had to build the right group by combining key members from both the visioning and learning product teams. From the visioning team, we paired a researcher—who leads our efforts to understand users and their unmet needs—with a designer who translates user insights into tangible ideas and solutions. These two “visioners,” as we like to call them, worked closely together throughout the project, with the designer heavily involved in research and the researcher heavily involved in conceptualizing new ideas.
We’ve learned that partnering closely with existing product teams, rather than working in isolation, significantly improves the quality of the vision and the likelihood that it’ll become a reality. In this case, we assembled a small cross-functional group from the learning team representing product management, UX, engineering, content partnerships, marketing, and business strategy. This group, combined with the aforementioned visioners, became our working team for this project.
“When the team embarks together on this journey, the sense of ownership is heightened. Alignment is achieved, and individual team member’s contributions increase in number and in relevance.”
Guillermo Krovblit, Product Manager
A visioning team is built with cross-functional collaboration in mind. Each team consists of leads from product management, UX, engineering, content partnerships, marketing, and business strategy.
Phase 2: Immerse yourself in the problem
Visioning allows us to both understand product usage, and the larger task of getting to know a new domain or industry. To explore a broad area like learning—which encompasses everything from math to makeup tutorials—we employed a 360-degree strategy for immersion, seeking insights from a variety of perspectives:
- Industry basics: We audited our existing product (YouTube) and evaluated the current options available in the industry to all kinds of learners. This gave us a solid foundation for the current state of our product and the industry at large.
- Experts and thought leaders: We met with members of the World Economic Forum to help us understand job reskilling trends; we were inspired by programs like Arizona State University and their mission to democratize education by making it accessible to everyone. Insights from these experts gave us a glimpse into the future, and provided great inspiration for new ideas.
- Empathy exercises: Team members enrolled in online courses to directly experience the learning process. These deeply personal experiences revealed insights, particularly on the challenges of staying motivated to learn.
- The learning ecosystem: In addition to understanding our audience of people using YouTube to learn something new, we broadened our research plan to include creators of educational content on YouTube, and employers too. We made sure to invest in research across all three user types to ensure that we designed solutions that created value for the entire ecosystem.
- The whole person: While data can be useful in understanding user behavior, understanding people on a deeper level requires getting out of the office and into the field. For this project, we traveled to different parts of the country to meet all kinds of learners and to better understand the person behind each individual user. We spent time with a donut glazer in Salt Lake City who was seeking a better career path but didn’t see any viable possibilities. He revealed that he grew up fatherless and lacked a professional role model to guide him. From stories like these, we learned that for many people the first step of their journey to learn something isn’t choosing a course, but the more fundamental step of discovering new directions for one’s life.
The team in transit on their way to a home visit in Salt Lake City. “In this early stage of the process, it can be tempting to jump right to solutions,” explains researcher and strategist Molly Needelman. “When working with new kinds of users and exploring new terrain, we quickly realized just how much there is to learn.”
Phase 3: Frame the opportunity
As in the immersion phase, it’s important to resist the temptation to jump immediately from observation to solution. First take a step back and map the broader opportunity space. For us, this began with team debriefs and storytelling after each home visit, and led to longer work sessions to identify patterns. Defining patterns helped the team identify the most important unmet needs among learners, creators, and employers, which we then translated into strategic opportunities, such as a learning journey map.
A workshop with the cross-functional team helps us to map out the learning journey and identify opportunities.
What started as a map made of colorful Post-its evolved into the more detailed framework for understanding learning and was foundational to our eventual vision. It helped us to translate and organize our insights into actionable moments that the team could design for. We also used the framework to prioritize the user needs we wanted to solve, and to focus our ideation sessions. By going through this process, we were able to zoom out and consider the entire end-to-end journey of learning, instead of focusing too narrowly on the experience of watching a single video.
Phase 4: Explore the possibilities (and fail early)
At this stage in the project some early ideas were beginning to emerge, but we pushed ourselves to stay broad and explore lots of proposals across a few key moments. The team generated a broad set of concepts, such as new ways to structure and organize learning content, features to help learners stay focused and motivated, and tools to help creators of educational content be more successful.
We created a wide range of ideas across the learning journey, and developed a card for each in low-fidelity sketch form.
Through feedback from learners, creators, and internal stakeholders, we identified a smaller set of concepts that felt promising. One idea that everyone could agree on was the ability to eliminate distractions—like recommended videos that don’t pertain directly to the topic at hand—when learning on YouTube. This began as an abstract concept, evolved into “study mode,” and ultimately resulted in video recommendations being hidden from the watch page (with the option to view them if desired).
“[This process] gives teams clarity around which ideas are most important to focus on and which ones to ignore—giving them the confidence to go build the product without second-guessing themselves.”
Yaakov Lyubetsky, Designer
A key part of this process was the philosophy of failing fast early in the process, with the use of low-fidelity wireframes and prototypes. We had many early ideas that we were excited about, but after testing them with users we learned that many of them weren’t any good. While humbling and at times discouraging, this approach ensured we focused our development efforts on the features and experiences that are best for our users.
Phase 5: Define the vision
Once we had fully explored the possibilities and validated our ideas with users, we pulled together the best elements from the previous four phases into a focused, cohesive, and tangible vision for the product.
The vision for the product pulls together the best elements from the first four phases into a single, multi-layered composition comprised of the following: 1. User insights; 2. Value proposition; 3. Vision pillars; 4. Signature concepts.
- User insights: The vision’s foundational layer is comprised of the rich user insights we gathered in phase two, such as the desire among learners for lifelong self-development and the desire among creators to showcase the rigor and credibility behind their work.
- Value proposition: Layer two articulates the user value of the product and is a good litmus test for whether the vision is clear and purposeful. In this example, the team defined YouTube’s value proposition as “a home for lifelong learning.”
- Vision pillars: The third layer defines the strategic focus areas for the product. We identified fueling personal development and preparing learners for success as two important pillars.
- Signature concepts: The final section represents the tangible expression of the value proposition and product pillars, inspired by user insights. The concepts defined here represent the major product initiatives we planned to invest in over the next two or three years, such as improved content organization, enhanced learning features, and creator tools.
Phase 6: Inspire action
Now it’s time to translate the vision into concrete, actionable steps by creating a long-term roadmap. Our roadmap was a bridge between where the team was and where we wanted to be in a few years.
We focused on defining the first and most immediate steps of the roadmap to create strong momentum towards our longer-term vision. This crucial work led to the 2019 launch of Learning Playlists, which divide a collection of videos into chapters around key concepts, starting from beginner to more advanced. This new feature also provides visual indicators of progress and hides recommendations from the watch page, allowing the viewer to focus on the lesson at hand.
Learning Playlists divide a collection of videos into chapters around key concepts, starting from beginner to more advanced—like this one by educator Hank Green (shown above). The feature launched in the summer of 2019 on both desktop and mobile devices.
The power of visioning
Thinking back to early 2018 when YouTube created a new product and engineering team dedicated to all forms of learning, it was clear that we needed to think beyond the small, iterative improvements that typically come from a versioning mindset—we needed to think bigger. Visioning enabled the team to make a big jump in their understanding of the product space, and it helped us define a series of leaps within the product experience. Learning Playlists was the first: this series of new features represented a significant improvement in the product experience, and provided a dedicated learning environment for people who come to YouTube to learn.
Visioning has had a tangible impact on YouTube’s product and users, and a lasting impact on the learning product team. “We met some really memorable people as part of the deep work on user insights early in the project, and those are the folks that stick with us and motivate us years later,” says UX lead Lettie Malan Assaf. “Each part of the final vision played a different role for the team. Signature concepts are still on our roadmap, and I often think about the vision pillars as we discuss new ideas and updates. It helps me remember the breadth of the space and be sure we’re focusing our energy on areas of value to users.”
Visioning inspires product teams to see new possibilities and think bigger, allowing our products to reach their full potential and truly improve the lives of our users.