CHAPTER ONE
CHAPTER TWO
- Rachel, a member of HeaRHfeltÂ
(potential volunteer on platform)
- Dr. Nadya, staff in charge of HeaRHfelt
(potential volunteer on platform)
- Dr. Nadya, staff in charge of HeaRHfelt
(potential volunteer on platform)
- Consensus from the 3 personal
friends interviewed
‍
Quantitative & Qualitative Research
‍
In addition to secondary research done on the market and competition, we used quantitative and qualitative methods to understand the problem and the various stakeholders involved in greater depth since we had access to them.
‍
Verify our hypothesis and confirm that this problem exists in NUS. Also, we need to scope out our users’ needs.
NUS Students who have attempted to make an appointment for counselling services at the university.
Accounts from a professor who receives frequent requests for mental health help from students. These students are often directed to the university's free counselling but still reach out to professor as they struggle while waiting for appointments.
‍
Snowball sampling of anonymous survey to personal networks of friends, friends of friends, interest groups and network in NUS
‍
- Anonymous
- Short
- Online
- Combination of multiple choice questions (MCQs) and open-ended questions
‍
135 NUSÂ students
‍
Our hypothesis was true with 84% of those who have made an appointment finding difficulty during the waiting period and having little to no one to turn to.
‍
Insights were used to develop personas and user journey maps to inform designs.
To get a better understanding behind the numbers of the survey, the problem, user needs and requirements, in-depth interviews were conducted with respective stakeholders. Because mental health is a sensitive subject, we decided to talk to 3 personal friends of the team who have turned to the university's free counselling. On top of conversations with students, we  interviewed 7 potential organisations that had potential staff and volunteers in the space for providing mental health support and would likely implement our platform, including the university's counselling services.
‍
The responses from potential organisations were varied due to the different organisational needs and the resources they had. Of the interviewed organisations, HeaRHtfelt, an interest group with passionate volunteers who aim to support peers who struggle emotionally from Raffles Hall (one of the university’s on campus housing) expressed great interest in the problem we were trying to solve.
Due to this insight from the interviews, we included a "flagging" feature in our platform whereby volunteers can flag a chat to supervisors should they need additional help.
Supervisors will be notified by email for urgent action such as addressing flagged chats.
Negotiating both stakeholder views, we decided to have both a sign up and an anonymous option for further testing and evaluation.
CHAPTER THREE
Those coming to Ora seeking temporary emotional support.
Trained volunteers from the organisation providing support to visitors.
Supervisors from the organisation overseeing volunteers and chat activity.
‍
User Psychology, Affinity Mapping, User FlowsÂ
Using primary research data and user psychology insights from secondary research, I led my team through an affinity mapping workshop and mapped out our user and product requirements. Finally, I drew up complex user flows for each persona and started designing accordingly.
I led an affinity mapping workshop to make sense of our research data and brainstorm product features.
Due to the limited time we had, I adapted the “$100 test” into a “time test” where features were prioritised based on development time (worked backwards) cross-referenced with the level of importance to respective users.Â
‍
To aid designing such a complex and interlinked system, user flows for each persona were drawn up according to their requirements and needs.
From our research findings, we saw the importance of anonymity for visitors. Anonymity was crucial to overcoming fear and social stigma associated to mental health issues and the assurance of privacy would lead to sharing more comfortably.Â
‍
However, many organisations felt that there might be spam from allowing anonymous chats and could not afford for that considering their limited resources available.
‍
From our survey results and qualitative interviews, convenience is a key requirement for visitors. The ability to chat on mobile is important as often visitors are in bed on their phones when they need support.
Through our conversations with those struggling with mental health, a comforting experience would be extremely key to making sure visitors feel safe on our platform and minimize drop-off rates.
‍
Because resources are limited and there will always be more demand than supply, I designed our platform to allow volunteers to talk to multiple visitors at once.
‍
To ensure that volunteers don't get overwhelmed, a clear and simple user experience was pertinent. As you can see in the volunteers' user flow (on the left), volunteers have one goal, which is to clear all chats in the left sidebar.Â
Qualitative interviews with potential volunteers highlighted the importance of being able to flag chats they can't handle to supervisors, informing our decision to incorporate this flagging feature into our platform.
‍
Volunteers can flag chats to supervisors should they feel like they are unable to handle any particular chat. Supervisor(s) will then take the necessary action.
‍
Volunteers can also check in real time whether any supervisors are currently online so as to streamline the flagging process, especially in urgent cases.
Drawing from insights found from our qualitative interviews with potential supervisors, it was pertinent that supervisors were notified of urgent action required when they are not on the platform.
‍
We learnt from our interviews with potential supervisors that they would need to be able to efficiently monitor chat activity and have control over problematic activities.
‍
Taking that into account, the supervisor side is designed to have a flexibility of actions.
‍
To ensure chats are replied in a timely manner, supervisors can re-assign volunteers or even join the chats themselves. Having visibility of who's online also helps in deciding re-assignment.
‍
For flagged chats, the supervisor can unflag them once they no longer require extra attention.
‍
To efficiently keep track of the monitoring of chats, a read/unread color change feedback is incorporated as well.
Supervisors also need to be able to manage staff. In this case, staff refers to volunteers and supervisors. Hence, the supervisor interface is designed to create new accounts, disable inactive staff accounts, edit roles such as promoting or demoting from volunteer to supervisor and vice versa.
CHAPTER FOUR
‍
‍
Little problems encountered aside from lack of response from volunteers when offline
Anonymous messages were mostly genuine, different from the large amount of spams that would have been expected
Concerns of being offline and getting a message → currently no notification if not on platform
User guides were helpful for staff onboarding
Question: What happens if a chat goes “unclaimed”?
-Â Feedback from Visitors
during the Pilot Test
-Â Feedback from Volunteers
during the Pilot Test
-Â Feedback from Staff
- Feedback from Staff
- Feedback from selected visitors and partner organisation
Many changes were made during the rapid prototyping process. To not overwhelm you, I would like to highlight these 3 major design changes after usability testings, 2 of which are interaction changes and the last one is a UI change.
The feedback from both visitors and volunteers highlighted the need for an interaction shift from majority cases being online and real-time to offline. This accounts for smaller volunteer teams and continuity of chats.
‍
Change 1: When a volunteer replies, anonymous users will receive a browser prompt while visitors with an account will receive a email notification.
‍
‍
‍
Change 2: To ensure visitors aren’t left hanging when staff is offline, an auto-message will be shown to them
‍
Change made: When offline, volunteers will be notified via email when there are new messages; when online but on another browser, there will be a browser prompt that will notify them of chat activity.Â
Feedback from HeaRHtfelt, usability testings and pitches with other organisations afforded for our discovery of various interaction flows organisations could take according to their needs.
‍
Ultimately, we want to partner with different organisations so that we can help as many people as we can.
‍
Change: Addition of “Auto Assign”
Chats are auto-assigned to volunteers with auto-reassignment for chats not replied in a set time period to ensure timely replies.
‍
Change: Choice of Visitor Login Page
Organisations are able to turn off the anonymous feature if needed
‍
Before the pilot test, we tested the originally dark user interface (UI) with selected visitors and potential partner organisations.
‍
Both visitors and partner organisations were not such a fan of the dark UI.
‍
CHAPTER FIVE
We had a really tight timeline of 2 months to produce a working demo-ready launch (inclusive of development). With more time, I would’ve refined the admin features and negotiate different stakeholder needs even more elegantly.
Negotiating different stakeholder needs is definitely a challenge. Conceptualising such a complex product requires a lot of thought and cross-checks with stakeholders, and oftentimes things get overlooked. Usability testing really helped in identifying the gaps, which were rectified efficiently and effectively.
Ending off with my favs, from our team win to publicity materials and illustrations!
Our team came in 1st-Runner Up for the STePs showcase which is an inaugural showcase held by National University of Singapore (NUS) School of Computing. During the pitch, many people really saw the need for such a platform, organizations and students alike. Our goal was to fill this mental health support gap and it was simply amazing to receive such validation for our solution.
‍
