Insightful — A UX Case Study
Insightful is an intelligent system for helping young adults understand more about their health and work towards suggested goals.
Before I started defining my problem statement, I needed to look for an actual real world problem that exists. From what I had heard form peers working within the medical space, health records are on their way to becoming digitised, and from my interest in technology, I knew that wearables are on the rise.
Looking into this area, I conducted some desktop research into wearables and digital health records, which then lead me to branch out into artificial intelligence, big data, and the strong link between all of them.
- The use of wearable technology for health and fitness tracking is growing exponentially
- Most people using wearables are between 24–35 years old
- People in this age bracket are generally more health conscious
- We are at the dawn of a new age of artificial intelligence
- Within the next 5 years, all our health data will be accessible digitally
- Artificial intelligence, in specific use cases, is being used to analyse and interpret large data sets
- There is opportunity to expand this to consumer applications
Hypothetical Problem Statement
How can we deliver health insights to conscious, young adults in a way that is understandable and not overwhelming so that we can help them proactively manage their health.
Before I started interviewing my target audience (young health conscious adults) I thought that it would be a good idea to get a high level overview of the problem, and what sort of scope I could be looking into — in other words, what was actually feasible.
This involved talking to a number of General Practitioners, who have an overview of many different facets of health, and how they work together.
What I needed to know
- What are the main reasons people come see a GP?
- Do people really feel they need to proactively manage their health?
- Should they be proactively managing their health?
- Can you predict disease and illness before it happens?
- What are the main causes for major diseases?
- People tend to only see a GP once they’re sick
- While people do need to be looking after their health, it is not generally done
- All major diseases have blood markers
- Exercise, cholesterol, food intake and heart health are all predictors of major diseases
After these interviews, I felt that I had enough knowledge of what to ask my stakeholders.
- Young adults are overwhelmed with information
- While they take steps to improve their health, it is misguided
- People who know more about their health will go to the doctors more often
I sent out an exploratory survey to my network in order to find any deeper problem areas in regards to managing personal health.
what I needed to know
- Do they currently have a wearable and use it?
- Do they track their exercise or food intake
- What is the correlation between exercise tracking, health and proactive health
- How confident do they feel about their own health
- What is the correlation between perceived health and actual health
I had 40 survey responses over the course of a weekend, from these results, I found that;
- 19 to 30 year olds are most affected
- 30% of people tracked their exercise
- 15% of people tracked their food
- People were generally not confident in their overall health
- 60% of people only go to the doctor when they are sick
- 25% of people get regular checkups
- 40% of people get general health checks
- Felt okay overall in regards to their health
- People who don’t track exercise also don’t track food
- People who tracked health also saw a doctor for general checkups
- People who saw a doctor regularly were less confident in how much they knew about their overall health
Before I went ahead with interviews, I had a few hunches from my results that helped inform the topics I wanted to talk about.
- People who are uninformed about their health are more likely to overestimate how healthy they actually are
- Health education is something that gets overlooked
- People who care about their health will make active steps to work towards something
I conducted a set of one on one interviews to get a feel for the deep issues that people had in regards to this problem, and find out what I didn’t know in order to go ahead with creating a product.
The interviews were largely unstructured conversations, which helped all my participants feel more comfortable in the space.
What I needed to know
- Why don’t they go see doctors often
- Why don’t they take steps to manage their health proactively
- If they do, what do they use and why
- What do they find easy/hard about managing their health
Any topic that was discussed I could dig deeper to discover the root cause and issue behind it.
At the same time as my interviews, I conducted a guided walkthrough with any participants that did use fitness apps, getting them to talk me through how they use the app and what worked and didn’t work for them.
- People respond really well working towards a set goal
- While there is data available from these apps, interpreting it is difficult
- They like to see how different parts affect the other, ie, how exercise affects sleep or heart rate
- Found that many of the apps they use lack any interpretation of their data
- It is up to them to conduct their own research to set goals
As part of my exploratory research I conducted a co-design session with a couple of stakeholders.
This was less of an ideation session and more of a look into the values and way people think around their health.
What I needed to know
- How do people want to be delivered their health information
- What do people consider health information
- What do people want to know about
- How do they want to action on their health
- What values do they hold in regards to their health
I facilitated this session with three participants. I explained the process, in which we would get five minutes to sketch out whatever came to mind around a topic area. Afterwards, we would explain to each other what we had drawn, and have a discussion around it.
“How might we learn about our own health in the future”
What we Discussed
Everyone had quite differing points of view, but what came across during discussion, was that they were all very valid points, that the other had not even considered. Some of the main points were;
- Having a conversion with your internal organs
- I need to know what’s wrong
- I need to know what I can do
- I need to know what I must do
- I need to work towards a goal
- I need to incorporate mind, body and spirit
- I need to understand the interconnectedness of my body
On the far left we had discussed technologies and machines that would analyse and display information to you directly.
In the middle, we discussed being able to talk to the inside of your body and ask your organs what’s wrong, understand what their needs are.
On the far right, we created a huge mind map of all the different aspects of our health, and how each one linked into the other.
During the entire design process I was clustering groups of information together, and identifying core needs.
I did a number of affinity maps during every step of the discovery process in order to condense huge clusters of information into something more manageable. Any outliers I kept to the side in case they could relate to another piece of research which I would conduct later.
My process was add, cluster, tidy, add cluster, tidy, Until there was nothing left to add.
The discoveries that I have grouped together for research were written up as story cards that define the needs of the people that I interviewed and did research with.
The main needs that I drew from the collective research were;
- I need constant and consistent feedback so that I remain accountable for my actions
- I need to understand what my body is telling me so I can help it
- I need to understand my mind body and spirit
- I need to understand what action to take so that I make the best decisions for myself
- I need to know how x will affect y so that I can make my own choices
- I need knowledge so I can feel confident in my own choices
- I need my information presented to me in a way that is easy to digest
- I need long term goals so that I stay motivated
- I need to see my progress so that I can work towards my goals
- I need to learn, not told what to do so that I don’t make the same mistakes in the future
Defining a Persona
I’m not really a big fan on personas, I don’t want to create an artificial model of a person because I would be projecting a lot of my own personality into it.
But what I can present are the facts, from my research, and from ethnographic studies.
- 22 to 30 year old (millennials)
- Surrounded by technology and comfortable with it
- Did exercise, track health
- Found understanding their health difficult
- Lack motivation and accountability
“A heath conscious young adult who feels lost in being proactive about their health, needs to understand their health and work towards a goal, but faces a lack of insight and direction.”
Based off of each insight, I did a brainstorming session in order to identify features that my solution could have in order to address each need.
For this stage, they were kept very high level, blue sky thinking. No idea is a bad idea kind of thing.
The Minimum Viable Product (MVP)
I used a grid map to define what would be the minimum viable product, sorting both features and insights on a map of easy to difficult, valuable vs not as valuable.
This way, the highly valuable and achievable options will be brought forward into ideation. The not as valuable, left behind.
I facilitated another co-design session with the same people that I had conducted the initial. This time, instead of drawing weird and whacky concepts for a future time, we were generating features and drawings based on the features I had defined as a minimum viable product.
For this session, I explained the process to my participants, that we were to sketch out as many ideas as we could in the allotted time frame; 5 minutes.
The ideas were to be centred around a feature that is to go towards a solution to my problem statement. So to give context, I brought up my problem statement and outlined the feature, we then drew for 5 minutes and afterwards, discussed our ideas.
I repeated this process for every feature.
What we learned
- An app is the best method of delivery for these features; everyone within this age group has grown up with technology and is familiar with apps
- Emojis are a great way to make something both relatable and fun
- Keep it simple stupid!
- Working towards a goal, but being directed with choice
- Not being told what to do is very important, but suggestions are okay
- Choice choice choice
- IFTTT style, if this then that happens
Information Architecture/Competitor Analysis
This was a really quick exercise in prioritisation of features and information delivery. In terms of aspects of the human body, what is most important to least important.
I was influenced by both my research insights, but also a competitor analysis of Apple’s Health app.
Wireframing / Dashboard
Okay, this is where things start going from a Waterfall into an Agile style workflow; so I guess, Wagile?
So my process here is to draw out screens that I think will work well as features within the app, and then prototype those individual interactions.
At this stage I did not have a clear flow for how the app was going to work, but individual features that I needed to get right.
These went through a lot of different stages of getting shown to stakeholders, tested, validated and iterated.
What’s this I’m seeing below?
Good question. That’s the main dashboard. How I want to present all of the information about the different interconnected systems. I’ve got 11 different concepts there on how I would both present and have people interact with that information. Ideas ranging from a body playground to an If This Then That style simulator, where you could adjust sliders and a little person would change as values were adjusted.
I also tested out the idea of a chatbot, and delivering information in a conventional way, like in Apple Health, as that a style that was familiar to people.
Prototyping / Role Play
I started things off designing my app as an iteration of Apples existing Health App.
Part of this was to get familiar with the design tools, Sketch and Principle, but it was also a great design exercise into getting into the mind of another company and developing a product from their perspective.
Why I did this
I chose this design method because it took a lot of the extraneous designs, (icons, looks and feel, structure, hierarchy) out of the equation and let me focus on getting the features to work just right.
This is a design method I used to use in Industrial design, in order to think about a problem from multiple points of view.
What did I learn
Well, I learned a heap about the design tools, but I also learned the importance of information architecture within the app, and how to split things up.
One thing I did notice about Apple’s Health app was that different sections were siloed, so you’re not able to see how one piece of information affects the other.
Also, while it is great to see everything together, ie, all your data in one place, it becomes quite overwhelming, and a lot of pieces of information might not really be relevant for you.
This problem will grow exponentially as more and more types of data are collected too.
I never did any usability testing on this prototype, as it really wasn’t my design.
Prototyping / UI / Usability Testing
In this prototype I wanted to not only test out the wireframe, but develop the aesthetic I wanted for the application.
I wanted to use emojis throughout as a fun way of describing without having to use too many words, and include a conversational tone within the copywriting.
- Easy to read titles
- Clean interface
- Conversational tone
- New information architecture
- Explanation of what you did and what you need to do
- Ability to add points of information to see how they work together
I ran a really quick usability test asking participants to check their vitals and add a new data set into the group.
- People really liked the colours, emoji and the copy
- The description was good
What didn't work
- People were used to seeing graphs and charts, so not being able to see that, they have to trust that the AI can pick up on those things
- There was no linking into how one thing affected the other
- They did not want to have to add data points manually
- There were no goals to work towards, no incentive to use the app other than to check things
Wireframing / User Flow
Going back to wireframing with my new findings from the last usability test, I wanted to look into how goals could be added and checked off.
The idea here is that goals would be suggested by the AI based on your health, and that you could tap to see more information about why these were being suggested to you, and make your own decision about wether or not you’d like to take on that challenge.
What in the heck is that?
That there is a really basic flow, tapping into the main page, which is a list of activities. Tapping into one of those activities would bring up more information about it.
Sliding over on one of them would confirm it, or add it to your list.
I also added an achievements page so that people can keep track of their overall progress and challenge themselves to push further.
So many Prototypes, and another Usability Test
I wanted to create a gesture that felt natural, sliding over, to bump a goal to complete it.
- Slide to check off activity
- View more information about why an activity is being suggested
- Choose which activities to pursue or not
- Understand what will happen if you do or do not do the activity
- See achievements over short and long term
- Suggested goals were received really well
- Again, emojis are a win!
- Seeing more information is great, but more graphics, less text
- Achievements section was a plus! People want to compare themselves to their own stats, it helps keep them accountable
What didn’t work
- The interaction was a little confusing to some, did not get the correlation between bouncing a bar and checking off an activity.
- No graphs were there, people want them!
- More emoji more more more
- More colours too!
I needed to prototype the entire user flow now, creating the entire app, as well as applying a new design aesthetic to it.
Apple human Interface vs Google Material Design
Reading both Apples and Googles design guidelines, I liked aspects of both. Being more familiar with Apples interfaces, and owning an iPhone, I decided to stick mainly to their guidelines, but incorporating a few details from material design, like the floating action button.
- Activity list, where you can choose from multiple options
- Activities will automatically check off once complete
- View completed activities
- View achievements
- See more information about each activity
- Simple layout
- New information architecture
How many times are we going to usability test?
— A lot.
- Up next
- Overall design
What can be improved
- More emoji
- Overlay page should be more direct, rather than a paragraph of text
- Interactions can be confusing
Opportunities for Future Features
- Information page should be more direct, if this then that style
- Adding more achievements to the page
- Try tabs instead of FAB, didn’t recognise icon, I think this was due to material design interface on iOS
- Achievement cards should be interactive, people were tapping on them a lot
- Can compare achievement to last month or the national average
- An on boarding tutorial process to guide on how to use the app and set context
- Revise the indicator buttons, too large and ugly
- Add a card flip style animation to the goals
Refining the Interactions
At this stage, I really wanted to get the interactions right. I did a few more rounds of usability testing and refinement — fine tuning the copy, interface elements, and animations.
The most challenging part of the application turned out to be the main interaction.
A series of tabs are presented on the main dashboard, and you are able to swipe horizontally on them to change between cards. I tried out many different variations to test what worked best for people; the last option was the most effective, but still not 100% effective.
My solution to this problem was to include animations in the on boarding screen.
The screens animate between three different images showing that the cards are able to be swiped over to change.
Turns out, quite a few people don’t read on boarding screens; a few people I tested skipped right through it without reading.
Some others went through the process, and then at the end mentioned they were distracted by the animations.
These were, however, a small minority of the people that I tested the application on.
Overwhelming consensus was that the onboarding screen and animations were very clear, and gave the app context.
The Presentation Arena
For the presentation, we’re given a small desk and presentation space behind to hang posters.
- Posters of major project (This one)
- Posters of past case studies
- Prototype on display
- Business cards
- Laptop with case studies
In addition to this, I’m adding a process diary, which is what you’re currently reading!
So let’s UX the whole process. In order to get everything ready for presentation, I did an Agile/Waterfall process in which I put every task that needed to be completed into a backlog. I did an MVP on those features to see what I should prioritise and get out of the way first. Anything important and easy was done.
Posters of Main Project
I really enjoy this process; whenever I make a poster for a presentation, I do a wireframe.
This way I have a very clear direction of that my posters are going to look like, where content is going to go, and what needs to be included.
So how do you even wireframe posters? Same way you’d wireframe an app or a website! Here’s a list of the posters I’ve got:
- Problem Definition
- Internet of Medical Things
- Demographic and Ethnography
- Business Model Canvas
- Facts and Figures
- Social Impact
- User Journey
- How it Works
- 4x Hero Posters
Past Case Studies
Okay so these are easy. I’m going to re-format my existing medium case studies, pop in some extra posters of documented artefacts. Done!
Prototype on display
So this is the part of the presentation that I feel like I wanted to put a lot of effort into.
I spent a long time thinking about how I wanted to display the prototype. It had to be interactive, which involved a phone. I wanted to change the phone, but didn’t want cables. So wireless.
This is the display I ended up choosing, which in turn meant that I needed to create an Apple Watch prototype.
That was such an effort that I wrote an entire case study about it.
Long story short, I taught myself Swift.
This project has been an invaluable opportunity to take on an issue that I have been passionate about for quite some time.
I’m also incredibly appreciative of the opportunities it has afforded me to learn and grow; speak to new people, and get out of my comfort zone speaking to industry experts and strangers.
I want to especially thank my family for supporting me through the project, both emotionally and financially, and my awesome peers Yolande Boulac, Nate Bamback, Leah Bayndrian, Akshati Shah and Emma Aylett who usability tested my app and gave amazing feedback.
Couldn’t have done it without you all!