Rising to the Challenge

Mental health is a sensitive topic. But imagine if you had to complete a UX design project on a digital product in that problem space. Plus, you couldn’t get face-to-face access to its users? What would you do?

This is the story of how we rose to the challenge of working on Uprise – a digital wellbeing program for employees of some of Australia’s largest enterprises.

The Client

Uprise is a three-year old startup founded by Dr Jay Spence. The platform is based on his PhD research and uses “stepped care” to support employees at varying levels of wellbeing – from surviving to thriving. This approach differs to traditional Employee Assistance Programs (EAPs), which only help those who are in crisis or already have a mental illness.


The Problem

The Client’s product manager approached us with two main problems:

  1. Onboarding and course completion rates were too low.
  2. Uprise needs to look and behave consistently across products and platforms (i.e the website and the App).

During our kickoff meeting with The Client, we agreed that in the context of a two-week agile project, it was only feasible to work on one problem.

Plus, it was neither logical nor sensible to work on a new visual design system until the first problem had been investigated. So our goal was to improve onboarding completion rates.

The client also expressed a desire for a chatbot. But existing research into how users prefer to interact with the Uprise platform and new technologies was not available. Nor were there any insights into the current problems. So, we advised that developing new solutions without doing the necessary research posed a significant risk to the business.


How might we improve the onboarding experience so that users of Uprise can make the most of what the platform has to offer?

Our general problem statement


The Approach

This agile project took place at Academy Xi (Sydney) during March 2019 and involved two sprints. I worked as part of a team of four UX design students. We managed the project using a Kanban board in concert with Trello. Our design process was collaborative in completing the following activities:

  • Research and discovery
  • User interviews
  • Heuristic evaluation
  • Competitor analysis
  • Affinity mapping and analysis
  • Forming insights and story cards
  • Ideating, wireframing and prototyping
  • User testing


My Role

My role involved significant contributions to desktop and qualitative research (including being a major contributor to the script for user interviews; and writing the complete script for user testing of the prototype), co-creation of the lean persona, ideation and design of new features (i.e. a new progress bar and “traffic light” emoji system); future opportunities (i.e. in-person seminars by the founder and a new revenue model); and co-preparation of the presentation deck of deliverables.

Sprint 1

User Research

The first challenge we faced was getting access to real users of the Uprise platform for face-to-face interviews. Although The client had recruited users for qualitative interviews, they had agreed to phone interviews. The Client had no prior knowledge of our need for face-to-face user interviews of users. Plus, many of the potential participants were not in the Sydney area. Nor could any local users make their way to our campus at such short notice.

Our solution to the problem was to conduct interviews over Skype or Facetime. Nevertheless, once we began to shift the expectation from a phone interview to a Skype/Facetime interview, most participants dropped out. We hypothesised that this could be due to the sensitivity or stigma around the subject of mental health.

I did some research around the topic of conducting qualitative interviews by telephone. Although this method is generally considered inferior compared to face-to-face interviews, a small but growing body of literature has demonstrated the value of phone interviews as a viable option for collecting rich narrative data around sensitive topics (Drabble et al. 2016).

So, to overcome to problem of many users being reluctant to participate in Skype/Facetime interviews (and to maintain consistency in our research methodology), we decided to conduct all user interviews (n=5) over the phone.


Recent evidence supports the use of telephone interviews for qualitative research around sensitive topics.


Heuristic Evaluation

We performed an heuristic evaluation of the Uprise platform to identify problems and opportunities that may not be discovered through user research alone. This involved evaluating the platform around criteria such as clarity, friction and distraction. Each member of the group went through the platform separately and we transcribed our observations onto sticky notes. Problems and areas of interest were identified using affinity mapping.

Competitor Analysis

We researched websites and apps in the mental health space to identify the direct and indirect competitors to Uprise. Positive and negative features of the various platforms were transcribed on to sticky notes. This enabled us to form “hunches” about which were the most engaging wellbeing platforms and identifying the reasons why.

Heuristic analysis (things we don’t like about Uprise) and competitive analysis (things we like and dislike about the opposition).


Insights

From affinity mapping and analysis of our user interviews, we discovered that users were not getting stuck during the onboarding process (i.e the initial well being and stress tests). It was at the next step of booking a coaching call with a psychologist (in order to complete the online modules) where users were dropping out.

The Client clarified that this is what he’d originally meant by “onboarding” completion rates. From our perspective, however, this was a “conversion” problem as it is a key point at which Uprise generates revenue.

After analysing data across our user interviews, heuristic evaluation and competitive analysis, we formed the following groups of insights and story cards:

1. What’s in it for me?

As a potential Uprise user:

I need to know the value Uprise might bring to me so that I can happily commit to the program.

I need a relatable and well explained score card, so that I can understand, gauge and improve my overall wellness.

2. Tests and scoring

As a potential Uprise user:

I need to know how my score is measured so that I know it’s credible.

I need clarity on the progress indicator when completing tests so that I can generate a valid result.

3. Coaching/Modules

As a potential Uprise user:

I need a coach that adds value to my progress in the modules so that I can confidently keep on track.

I need to access the modules without booking an initial coaching call, so that I can help myself, learn at my own pace and decide later.


After presenting these insights at the first client showcase, we agreed on the deliverables for Sprint 2: a persona and customer journey map. And if time permitted, a prototype of some new features based on our insights.

We reiterated that to produce these artefacts, we needed face-to-face contact with real users.

Affinity mapping showing details of key clusters . The cluster on the left indicates that users could not recall having difficulties with the initial wellness tests during onboarding. The cluster on the right highlights that users were not ready or comfortable with being forced to book a coaching call.


Sprint 2

The Client was unable to provide face-to-face access to the required users for our Week 2 sprint. But this was not confirmed until Tuesday afternoon, which meant we had to revert to our backup plan – create a lean persona, a user flow of the current platform and a prototype.

Meet Rebecca

Based on our user interviews and some guerrilla research using colleagues, staff and friends, we created Rebecca – a lean persona.



User flow

Without the ability to conduct face-to-face interviews with real users of Uprise, we were unable to create a customer journey map. So, two members of the group mapped the current user flow, which highlighted the inherent complexity and confusion. Producing this artefact confirmed many of the hunches from our heuristic evaluation.



Rapid Ideation, Wireframing & Prototyping

Given time restrictions, we were unable to address all of the insights that we’d uncovered. So, we prioritised the most important groups of insights (i.e. groups 1 and 2 above) using “dot voting”.

Next, we ideated around how we could design to these insights using 6ups, sketches and wireframes.

Within a few hours, we produced a collaborative prototype (using Figma) containing new and revised features with the overarching goals of improving clarity, simplicity, navigability and relatability. The key new features of the prototype included:

  • Revised hompage
  • Overview page
  • Wellness check progress bar
  • Wellness test “emoji traffic light” scorecard and graph
  • Simplified dashboard vs. Current dashboard

Overview Page and Wellbeing Check: from wireframe sketches to prototypes.


Emoji “Traffic Light” Scorecard: from wireframe sketch to prototype.


Simplified Dashboard: from wireframe sketch to prototype.

With user testing scheduled for the next day, I volunteered to write the script, while another member of the group refined the prototype that we had built earlier in the day.

Usability Testing & Analysis

We tested the prototype with four potential users. Unlike traditional Employee Assistance Programs, Uprise doesn’t simply target users who are in crisis. So, although we weren’t able to do detailed screening of participants, our sample contained a mix of users experiencing varying levels of workplace stress (from high to low).

What worked?

Affinity mapping and analysis of our user testing data revealed:

  • The new overview page was clear to users and met expectations. Knowing all the steps before starting tests was understood. Plus, most users enjoyed and valued the video in this context. Originally it was placed elsewhere on the website.
  • Users understood new progress bar and could see which step they were at in the process. Plus, it was consistent with the format of the progress bar in the left-hand sidebar.
  • Users found the new radio buttons easy to use.
  • The new emoji “traffic light” scorecard resonated strongly with users.
  • Users found the comparison graph easy to interpret. The simplified version of dashboard was preferred by most users.

Affinity mapping from usability testing of our prototype.


What needs improvement?

  • The homepage needs further iteration to improve clarity of the value proposition and to set user expectations.
  • Users are confused over what is a “module” and what is a “skill”.
  • Users want to unlock and complete modules without the guidance of a coach.
  • Consistent visual branding and new design language.


Future opportunities

Based on our research, analysis of data and ideation, we identified the following opportunities:

Real life onboarding – examine the opportunity for presentations and on-site introductions to the platform for employees in workplaces. If employees are made aware of the platform and its benefits prior to suffering any mental health issues, they are more likely to use and understand its value now and into the future.

Plus, in-person events provide an opportunity to conduct research and collect feedback from employees via surveys in order to iterate and improve the Uprise platform.

Ideation produced two similar non-digital solutions focussed on in-person events.

Build community – examine the option of a forum or message board (moderated by coaches) to enable people to connect with one another. Building community is a central component of many successful apps.

Paid modules – primary revenue could be generated through access to modules (which is scalable), with coaching calls as an upsell. Most users expressed the desire the work through the modules without a coach. Competitive analysis also indicated that a business model based on unlocking modules is sensible and scalable.

Enhanced assessment – more detailed assessments including personality (not just mood, energy, engagement).

Further researchthere is a risk that the results of our project are not applicable to real users. Further research and testing with real users is required to validate new features and ideas. This includes determining the ways in which users prefer to interact with the platform and new technologies, such as chatbots.


Client Satisfaction

I think you’ve all done a great job at this project – despite all the limitations and ambiguity. The insights you have given us are excellent and we’re putting them into our backlog immediately.

Chaitanya Sharma, Product Manager, Uprise.


Acknowledgements

Thank you to Chaitanya Sharma from Uprise and Robert Williams from Academy Xi for giving us the opportunity to work on this project. Thanks to my fellow team members for digging in despite the difficulties: Esther Gadate, Akshati Shah and Versha Karwal.

Team Uprise.

Reader Interactions