Google Design Challenge
Feb 2020
Empowering Communities by Facilitating Mentorship
A fast, interactive mobile experience for the discovery and maintenance of mentor-mentee-relationships.
Problem & OpportunitySolutionKey FeaturesProcessStrategyDesignPrototype & TestPitchReflection
Intent:
Design Challenge
Role:
Strategy, User Research, UI/UX, Prototyping
Tech:
Figma, Proto Pie, Adobe CC
Duration:
1 week
Many undergraduate students don't have a mentor or don't know that mentorships are available to them. However, mentors provide a lot of benefits for new students who may struggle to adjust to campus life. Through VCU Connect, a new mobile experience that easily let's new students to connect with mentors, community is strengthened and mentorship is finally facilitated.
The Ask – Prompt #2
Your school wants to strengthen the community by encouraging experienced students to connect with new students and help them adjust to campus life. Design an experience that allows mentors and mentees to discover each other. Consider the needs of both mentors and mentees, including how someone may become a mentor and how to connect mentors to mentees.
Background
A mentorship provides numerous advantages for both mentor and mentee. Research shows that mentees generally perform better in their programs and after they get out of school, according to W. Brad Johnson, PhD. In addition, they gain practical advice, encouragement and support, develop their communication, study and personal skills, among many others. Mentors, on the other hand, can develop leadership and management skills, reinforce their own study skills, and set themselves apart by enhancing their CV, to mention a few.
Solution
Introducing "VCU Connect", a mobile experience that helps new students to better adjust to campus life by providing tools to discover mentors with common interests and facilitate communication. VCU Connect is personalized, transparent, and makes mentees autonomous.
Please wait while the video is loading. It will autoplay.
The Process
In the following, I will throughly explain the process of arriving at my final solution. Alongside, I will give thorough explanations about my decision-making for several critical decisions.
The Process: Strategy
Defining and Understanding the Problem
Scope
While "new student" could potentially have many meanings (undergraduate, graduate, PhD) I would like to focus on freshmen (for personas, user journeys, etc.) who connect with more experienced undergraduate students (seniors) for the sake of the narrative. However, the final solution can be used by all students interested in finding a mentor and all experienced students who are interested in becoming a mentor including master's and PhD students.
Secondary Research
To get a better idea of the existing infrastructure and the problem I am ultimately trying to solve, I started my discovery phase with secondary research.

Key Facts about Existing Mentorship Programs at VCU
Existing mentorship programs are not widely-known and are differently organized for each department of study which results in unequal opportunities.
Key Facts about Rams Connect
Rams Connect connects students with student organizations and helps finding events but it does not pair individual students with experienced mentors or vice versa.
Third-Party Platforms to Connect Students

Reddit

Facebook Groups
Third-Party platforms are not an efficient tool to connect new students with experienced students due to their broad spectrum of information.
Initial Research Assumptions
Primary Research based on Initial Research Assumptions
Through 1-on-1 interviews, focus groups, and a survey, I sought to find out more about mentorships at VCU and back up my initial research assumptions.

Research Results
Below are my key insights from a research survey among 42 VCU students that are either freshmen or remembered the time when they were.
Quotes
Below, there are the most helpful insights from my 1-on-1 interviews and the focus group.
"I knew nothing about mentorship programs as a freshman and now I am a senior. This is really my first time being introduced to the idea so I think marketing these programs in dorms could be helpful."
– Survey Respondent
"I feel like the organization behind many programs are not on top of it, that's why there is not much engagement."
– Catherine, 1-on-1 Participant
"As a freshman the closest thing I had to a mentor was my RA [Resident Assistant]."
– Kyiala, Focus Group Participant
To conclude, nearly all of my assumptions have been proven by research, except the fact that inclusion is not as much of a problem as far as finding mentorships is concerned. Unequal opportunity is much rather a problem of bad organization which can easily be solved with a centralized, uniform solution.
Affinity Map
I used affinity mapping to organize all research findings and determine areas of opportunity.
Opportunities
After I organized my research, I determined the following areas of opportunity:
Personas
To better understand the users that would utilize the experience I created Personas for the primary user (new student / mentee) as well as the secondary user (experienced student / mentor).

Primary User
Secondary User
Competitive Analysis – Finding / Matching People
To get an understanding of how existing experiences solve the problem of finding or matching users, I conducted a competitive analysis.

Tinder – Matching Process
Pros:
Fun matching process

Cons:
Focusing on looks
Highly iterative process

While matching with a mentor is based on certain criteria like program of study, age, level of commitment, etc., mentees will not be as picky as on a dating platform and they most likely will keep a mentor for a long time. However, mentees should be able to choose from more than one recommended mentor, so it could provide a more playful onboarding experience.

LinkedIn – My Network Page
Pros:
Compact overview over numerous people

Cons:
Not much information visible without tapping to go to the detail view

As new students will have to choose their first mentor from multiple suggestions, it would make more sense to show them all at once instead of letting the user swipe (c.f. Tinder) and make a final decision without even having seen all suggestions. Therefore, an arrangement like this could make more sense for the onboarding experience.

Recommendations such as the ones on LinkedIn could provide a way for mentors to get rated so mentees know whether a person is holding true to their provided information.

Facebook – Search
Pros:
Efficient searching and filtering

Cons:
Adds complexity

I could imagine something similar as a search filter to find mentors more specifically.
Feature Brainstorming based on User Needs
With my research in place and through inspiration and conclusions from the competitive analysis, I started to brainstorm what features need to be in place that help the users achieve their goals taking into account the pain points of my primary and secondary users.

During that process, I asked myself various questions including the following:
If you match with a mentor once where does the experience provide you continuous value?
When the experience provides sufficient communication tools, it could continuously provide value as an inner university circle of helpful experienced people. There should be some sort of personalized recommendations for mentees about people that might be interesting to them, e.g. on a community page. The app should encourage new students to spark the conversation with multiple mentors.
How do you become a mentor?
Could be through a sign up process that may include an application. Instead of an application it could also be as simple as uploading proof of your grades to show commitment to your studies.
Should every mentor go through an extensive application? Or is that just a plus?
Submitting an application could increase the chances for mentors to get requests and give them a verified badge, but some mentees prefer low-level messaging or casual one-time coffee meetups, as proven through user research.
How does the app experience differentiate between mentor and mentee?
For mentees it focuses on finding people (mentors) while for mentors it is focused on managing people (mentees). At its simplest form, it could be fundamentally the same and only slightly rearranged to shift focus from discovery to management.

The outcome of the feature brainstorming is as follows:
Feasibility Matrix
Due to time constraints for this project, I created a feasibility matrix to quickly determine features with the highest feasibility and impact in order to prioritize where to put my area of focus.
As the feasibility matrix shows, I focus more on community and communication features and deprioritize search filters and tools that facilitate meetups.
Core Features (for Mentees)
Connect
Connect mentees with mentors

Communicate
Private messaging between mentee and mentor

Notify
Inform user about accepted requests, recommendations, etc.

Personalize
Change personal information, skills, and more
Core Features (for Mentors)
Manage
Manage mentorships, change personal information, skills, discovery options, and more.

Communicate
Private messaging between mentee and mentor

Connect
Connect with mentees

Notify
Inform user about open requests, recommendations, etc.
The Process: Design
Wireframing: Thinking Through the User Flows
Onboarding
Key questions I asked myself for the onboarding process include the following:
Where to ask whether the user is a mentee or mentor in the sign up process?
Initially, I was thinking about creating a label along the lines of "Not a mentee? Login here" that takes you a mentor login page given the standard login will be for mentees as this is the primary user group. However, after thinking more logically, you only need to be asked once and next time you login the system knows whether your account belongs to a mentor or mentee account. Hence, during the sign up process there is a "I am a - mentee or mentor" section with radio buttons to choose.
If you want to be both a mentor for some students and a mentee, then you would simply sign up again as a mentor or mentee. If the system detects you have another account with the same email, it will create a new sub account and next time you login you will be asked whether you want to login as mentor or mentee.
How does the sign up process for mentors differ from mentees?
Both sign up processes are fast! For mentors, the only difference is the omission of the very last step, choosing a mentor. Other than that, both feature the ability to upload proof of the GPA in order to show commitment to studies which may be a relevant selection criteria for some users.
How to design the initial mentor selection?
The idea here is to have alternatives and exactly 1 choice to make. Every mentor can be fully reviewed before finalizing the decision and this process can technically be skipped and resumed later.
Connect Page
The following image shows a low-fidelity wireframe of the Connect page.
After a mentee has found his or her first mentor, this is the page that should play a signifiant role in providing continuous value.
Key questions I asked myself about this design include the following: 
How to provide continuous value?
In my opinion, the experience should be capable of more than matching mentees with mentors and then never using the app again. Having multiple mentors is beneficial for personal growth according to an article from Santa Clara University. That being said, the experience should continuously recommend interesting people with diverse expertise and learn from your in-app behavior and provided information, who might be an interesting fit. At the highest level of technology, machine learning would be an ideal implementation for this.
How to visually organize a page for continuous value?
In my design I wanted to create a perfect balance between highlighting certain users more in detail but also being able to go through a bulk of users more quickly. Therefore, I created a "Recommended For You" section at the top that showcases mentors more in detail while there is also a "Popular New Mentors" section that only features one line of text as detail besides current studies, name, and age, allowing users to scroll through more quickly. After this section, there might be another more detailed section fro highly skilled mentors to keep up the balance.
What's the point of the verified badge?
The verified badge is supposed to show that users uploaded a transcript to verify their GPA. A badge is not the equivalent of a superior GPA / skill / expertise, but it shows that the user is more serious about this platform and his studies. Having a verified badge also allows users to have custom qualifications such as "This mento is doing exceptionally well in her studies.".
Message Page
The following image shows a low-fidelity wireframe of the Message page.
The Message page is a very simple page with different in-app conversations for private messaging.
Key questions I asked myself about this design include the following: 
How to differentiate mentor conversations from less relevant ones or emphasize them?
There are several ways to answer this question. I ended up deciding for a simple mentor badge. The reason for this is that, while talking to many mentors is encouraged, there will not be as many conversations as in other messengers meaning that a badge is enough to keep it organized. Second, if someone is your mentor, you will naturally stay in touch more which pushes the conversation to the top again.
An alternative could be that mentor conversations are pinned to the top and other conversations are below. A third alternative could be similar to Instagram's messenger tabs "Primary" and "General" which they introduced as it was unorganized when receiving a lot of messages which is the case for some business accounts. However, as I have said this would be unlikely for this experience.
Notification Page
The following image shows a low-fidelity wireframe of the Notification page.
The notification page is also rather simple and shows the the user different notifications such as accepted requests, new messages, etc. Key questions I asked myself about this design include the following: 
Is an entire tab bar page dedicated for notifications really necessary?
A notification page is necessary as it provides a history of previous interactions and helps to provide continuous reasons for the user to open the app. It is not necessary to put the page in the tab bar but due to the simplicity of the app it is the best place to put it as the user can most comfortably access it.
What should the user be notified for?
The most important notification is when a mentor accepts a mentorship request or when the user receives a message from a mentor. Other possible (less relevant) notifications could be that recommendations have been updated or that interesting new users joined. Again this is a way to provide the user continuous value.
How can the user have quick control over notifications?
To ensure this, I added a 3-dot / more button to each notification cell where the user can quickly manage / turn off specific notifications.
Profile Page
The following image shows a low-fidelity wireframe of the Profile page.
Key questions I asked myself for the Profile page include the following:
What value provides the Profile page to the user?
It's the place where all the profile info can be changed and mentorships can be managed. It also gives an overview over open requests that mentees have sent to mentors and they might easily cancel them there.
Profile Detail Page
The following image shows a low-fidelity wireframe of the Profile Detail page.
Key questions I asked myself for the Profile page include the following:
What should be part of the Profile Detail page?
The Profile Detail page should provide the user more detailed information about the expertise and personality of a mentor to give a first impression whether they might be a good match or not. It should communicate the level of expertise and commitment, interests, a bio, recommendations, etc.
Where should buttons be placed to request a mentorship or message users?
While they are placed differently in this low-fidelity wireframe, an ideal place would be at the very top to ensure quick and easy access.
How to make communicate the users level of commitment, expertise in his or her studies, etc.?
As a solution for this, I thought of qualifications that are automatically generated and are placed at the very top of the profile. A stopwatch icon stands for high commitment, a cup stands for expertise in studies (based off an uploaded transcript), and a star means the mentor has received recommendations from previous mentees. Recommendations are similar to LinkedIn and describe a mentee's experience with a specific mentor.
The Process: Design
High-Fidelity Screens
Onboarding
For the onboarding experience, not much has changed compared to the low-fidelity screens. The only difference is the checkbox which is intended for very new students who do not have transcript just yet.

For the mentor sign up process, the onboarding experience is quite similar except for two changes. First, for mentors there will not be the discussed checkmark as experienced students definitely know their GPA or have a transcript. Second, the last screen about requesting a mentor will not be part of the onboarding.
Connect Page
For the final Connect page design, I swapped the two lines of text in each card with icons about interests or qualifications. It didn't make sense for me how to fill two lines of text without making the user put in two custom lines of text for their highlight cards. Alternatively, generating text that describes interests or qualifications seemed less benificial and more complicated. Icons make much more sense.
Message Page
Notification Page
Profile Page
Profile Detail Page
Differences for the Mentor In-App Experience
As previously stated, for the mentor (secondary user) in-app experience, the focus of the application shifts from discovery to management of requests or existing mentorships. Therefore, the tab bar at the bottom is slightly rearranged to conform with the most value-creating features for mentors.
From the Manage page, mentors can easily have an overview of their current mentees and open requests. They have the ability to jump right in a conversation with them and will also see new messages as a badge.
In addition, mentors have the ability to hide themselves from the Connect discovery page or search where mentees might find them. This does not affect existing mentorships and it is ideal when a mentor decides that he or she has enough mentees or wants a break from getting requests.
The Process: Prototyping & Testing
High-Fidelity Prototyping, Interaction Design, and Testing
To prototype, demonstrate, and test my concept with users, I chose the prototyping application "Proto Pie" which I believe is a very powerful prototyping tool.
Reflection
This was a fun problem to tackle and I am overall impressed how much I am able to achieve in such a short time. It reinforced my decision-making skills which are crucial under a project with strict time constraints such as this one.

Thank you so much for taking the time to read through my execution of the Google Design Challenge.
Team
Dominik Hofacker
Experience Designer