Product Designer

Medical diary


Redesigning medical tracking

One of the core fundaments of Melon Health's mobile application is the ability to track how you're doing on the go with our native Android and iOS application.

The purpose of tracking is the gain insights and see trends that help the user gain insights about their chronic condition.

The more often the user tracks, the better the insights.


Your health to do list

The first concept was to make a to do list. By tracking, the user is rewarded by showing a visualisation of their progress.

Screen Shot 2017-11-26 at 15.29.09.png
Screen Shot 2017-11-26 at 15.29.15.png

Testing assumptions

In order to design the new medical tracking (formally called Diary), it was important to gain insight into how our users were currently tracking. I investigated the Google Analytics to help me understand user's expectations and behaviours. This understanding allowed me to update the design without forcing the existing users to have a steep learning curve.

Pasted image at 2017_10_18 08_54 AM.png

Visual explorations

I did a few quick experiements and explorations of how to represent a to do list in a visual and non-threatening way. In this case, I tried to represent the user's progress graphically in the same place where the call to action was to track.

Ultimately, this was muddling too many features in one screen. The most primary function of this screen is to allow the user to enter values about their health, and hiding the call to action in a graph was a disservice to the screen's goal.

new filled copy.png
new filled.png
new filled copy 4.png

Rewarding the user for tracking

Taking a step back, the most important thing was the user would input how they were feeling at least once a day.

In the second iteration of the interface, the user has several large calls to action allowing them to input various values (such as mood, energy, food diary, symptoms...)

When a user has inputted their value for that day, the block changes into a visualation of their progress as a small reward and source of navigation.

Screen Shot 2017-11-26 at 15.29.54.png

For some simple inputs, such as happiness, it was possible for the user to directly enter the value on the screen itself. Some other metrics, such as entering a food diary, were more complex and warrented the user being directed to a detail screen.

Screen Shot 2017-11-26 at 15.30.16.png

Entering complex metrics

Each detail screen had it's own set of challenges. For example, users measuring blood glucose for diabetes needed to easily enter their blood glucose around five times per day. Each entry of blood glucose also had optional other fields that helped the user gain greater insights into their blood glucose trends.


Templates & idiosyncrasies

As much as possible, I tried to adhere the detail screens to a template to reduce development work and user's learning curve. However, each metric tracked had it's own logic and requirements so some deviation was necessary.

Screen Shot 2017-11-26 at 16.08.24.png

Visual style

One of the most important elements of the redesign was an attractive visual style. As people were putting in rather senstive data, it couldn't feel goofy or untrustworthy. However, we wanted to continue our brand style of approachable, human, and non-clincial design.

The end result was an interface that was rather vibrant and playful, without being too silly.

Screen Shot 2017-11-26 at 16.08.42.png

Weak points, iterations

It was important to reflect upon what could be improved in the new interface. While it was very visual and graphic, it made the screen quite long if a user was tracking many different. This also somewhat lost the to-do list feeling, as a normal screen could only show two or three items at a time.


Trials, more iterations

To address the problems in the new design, I quickly tried a new concept that was less visual and more of a traditional to do list.

It's definitely simpler but might lose a bit of the context that the first iteration provided.


Next steps

This project is still in progress, so the next step will be getting some current and new user's feedback in a series of user tests.