Bundling all components of mentorship connection into one convenient package.
I led the design execution, including the lo-fi, med-fi and hi-fi prototypes. In the first stage of the project, I worked with three other team members to do needfinding research and brainstorming. Then we're divided into 2 subteams, where I worked with Kayla to implement a design solution to the problem that we identified in the first stage. Here, I took the lead in constructing and testing all three different prototypes. Virtually all of my design decisions ended up in our final solution.
After the class ended, I took a week to redesign the UI - polishing the UI we did in the previous 10 weeks with some details that didn't end up in our class submission due to time constraints.
How might we improve the course selection
experience for students?
Understanding the current student ideologies around course enrollment showed systemic areas of improvement rather than solely problems with SimpleEnroll or Axess (Stanford's enrollment platforms).
Customize your mentorship
Easy, quick account sign up and orientation
Allows you to select your needs
Finds mentors based on your needs
Select your mentors
Shows suitable mentors based on
Shows mentors' bios
Safe & secure communication
In-app meeting scheduling
Meet with your mentors
Conveniently schedule a meeting from
within the app
We started out our needfinding process by splitting up our student actors into two groups: Course Enrollment Novices (New Enrollers) and Course Enrollment Veterans (Experienced Enrollers). After several interviews with each group, we gained the following main insights:
Some experienced users had to use a default/dummy schedule provided by their major as a base and then alter it accordingly.
Transfer students have to plan ahead very early in order to avoid red tape while lacking lack data about next year's classes.
Inexperienced users think official academic advising is not as helpful as peers’ recommendations.
Unpacking needfinding data
Journey map - Inexperienced user
Journey map - Experienced user
Design Idea 1: Non-traditional Course Metrics
We created a course exploration tool that provides students with information regarding unconventional matrices that are unavailable on current course exploration sites.
A class posting with non-traditional metrics
Comprehensible data with visual elements.
Offers students insights into a wider array of aspects of a course.
Aggregated data is not personalized enough for individual students.
Design Idea 2: Mentoring Forum
We recruited people with majors that we ourselves are familiar with. The reason for this is that one of our members will act as the “mentor” in the prototype so we need to be familiar with the major ourselves in order to give advice through a messaging platform.
Screenshots from one successful interaction
Users report feeling more confident and certain about their coursework decisions.
Students are more comfortable learning from their peers than from someone in the administration.
An “unofficial” peer advisor might have less expertise or required experience and thus be less helpful in some situations.
Design Idea 3: Course Tinder
We tested to see what students think about a simpler and easier process for building the ideal schedule by continuously comparing a few or two classes at a time. We used the favorite picker off of Github to quickly make a ranking tool. This ranking tool would essentially allow us to easily rank classes based on comparisons between many sets of two classes each.
The first iteration of “Class Ranker”, included are 19 undergrad open option computer science classes available in Spring 2020
The picker gives suggestions that are accurate to users' interest.
Is not personalized enough for individual users.
Users find the picker's suggestions irrelevant.
After carefully considering all the pros and cons to each design idea, we decided to focus on the second approach: mentoring. We started out with creating initial concept sketches.
Low Fidelity Prototype
Confusion about prompts' meaning
The prompt "What best describes you?" on the onboarding screen for mentee's class year is too generic.
"Next" buttons (>) were not intuitive
Users had a hard time finding the > on the right edge of the screen.
Medium Fidelity Prototype
Digital grayscale mock-ups
Button hierarchy was not super clear
Buttons were not properly weighted based on importance.
Navigating pages for scheduling a meeting was clunky
Users had to go back to message history to schedule a meeting.
The Mentor cards were well-presented
Users liked the layout of the mentor cards view.
High Fidelity Prototype
We conducted multiple RITE tests of our final prototype and got a lot of useful feedback.
I want more unique, editable text to truly learn about the mentor I am seeking help from.
People will not always have the same needs.
It is hard for users to keep track of what they have chosen.
I would like to see matching options after typing a bit instead of scrolling all the way.
From these feedback, we came up with the following design recommendations for our app:
Provide editable textfields during onboarding for people to write their personalized bios.
Allow users to input custom requests.
Have a set of tags at the top of the screen that allows the user to see what they chose.
Support dropdown for manual type in during onboarding.