DCO.ai
DCO.ai's product TagStream is a language and image understanding tool that populates tags from concepts or keywords in a video.
​
Our objective was to redesign their product for a different use case.
​
Roles: Teammate, UX Designer, User Researcher, Content Designer
Problem
TagStream is currently being used by media professionals, but the stakeholders want to target the academic sector.
​
Determine if TagStream would be useful for students, then redesign the interface for learning.
Hypotheses
We will discover that students adopt similar patterns of behavior when interacting with educational media.
​
TagStream will help students locate specific areas of a video and provide students with a general understanding of the video.
Opportunities
TagStream can be used as a study tool and help students manage long-form videos.
​
Provide informational context about a video's topics through timestamps, video segmentation, and related tags.
Tools & Methods
Design: Figma, Miro, FigJam, Personas, Sketching, Wireframes, Task Analysis, User Journey, User Flow, Prototyping
Research: Semi-structured interviews, User testing, Competitive analysis, Literature review, Qualitative coding, Affinity mapping, Concept testing
Solution
Redesign TagStream to be organized in a way that simplifies tags into manageable chunks for findability while also providing students with an opportunity to view the video granularly.
Old Design
New Design
Research
Questions
How do people feel about learning from educational videos?
How do people use video interfaces when watching educational videos?
What are some challenges people encounter when learning from videos?
What about videos do people focus on when learning?
Methods
01.
Literature Review
Purpose: Confirm that videos are an effective learning tool and learn about validated design decisions that best optimize learning.
​
Discovered:
-
Videos should signal key concepts
-
Splitting content into chunks is recommended
-
Interactive features give students control
​
02.
Competitive Analysis
Purpose: Establish an understanding of the problem space and how other products assist with online learning.
​
Discovered:
-
Many audio transcriptions exist to help with note-taking and recordings, but there is an opportunity for video tools like DCO.ai that integrate visual note-taking as well
​
03.
Semi-structured Interviews
Purpose: Allow meaningful investigation participants' attitudes towards educational learning.
​
​
Discovered:
-
There are benefits to learning from videos but elements of classroom learning that videos lack
​
A Deeper Look Into Semi-structured Interviews...
Participants
We wanted to target educational use cases of videos where the learner has high motivation to learn.
so we tailor the product to multiple forms of learning, and not only formal education
When we say self-learners here, we mean people who are taking it upon themselves to learn a skill that will advance their professional or academic career
Note: Participant has okay'd picture for portfolio purposes, but I blurred the image to respect their privacy.
TagStream's interface is viewable because I used their service to familiarize myself with the product during the research phase.
We interviewed 11 students and self-learners
Community College
Undergraduate students
Graduate students
Self-learners
I recruited 8 of the 11 participants, interviewed 4 of them, and coded 6 of their interviews.
Affinity Mapping
After my team and I finished coding our participants, we worked collaboratively to identify themes across each interview.
Findings
Long-form videos may deter students from watching the video.
Students use interface features to avoid the length of a video and find only what they are looking for.
Design Recommendation: Findability
Showcase the density of frequently marked tags.
Visuals can communicate difficult concepts, ideas, or explanations.
Visual representations are effective in helping students understand what they're learning.
Design Recommendation: Video Clip Tags
Allow screenshots to be used as tags so students can tag memorable visual moments.
Students rely on and expect common video features.
Videos allow students to control the pace of their learning.
Design Recommendation: Findability
Showcase the density of frequently marked tags.
Students seek interactions with peers even through online learning.
Collaboration and class discussions are classroom elements students miss.
Design Recommendation: Collaboration
Students should be able to interact with each other.
Design
Content Design
