SELF: A Quantified Self Dashboard

Defining the Problem

There are a generally a few different types of people when it comes to exercising:

  1. Athletes and fitness fanatics
  2. Most people, who want to be fit and healthy but perhaps don’t do everything they could; and
  3. Couch potatoes who don’t do much more than walk to and from their desk all day

For those in every category, there are now more options than ever to track a myriad of personal data. The availability of so much data has led to a movement known as Quantified Self, the goal of which is self-improvement through quantified and analyzed self-knowledge.

One of the biggest limitations of the Quantified Self movement, however, is the difficulty to combine and draw big-picture insights from so many different data sources. This project, SELF, aims to solve that limitation through a dashboard that unifies and analyzes your data, tracks your goals, shows you data trends and, most importantly, gives you deep data insights into how to improve yourself.

My Design Process

I followed user centered design and lean UX processes to make sure that all of my design decisions were validated by user research, empathy and feedback.

Personas

I created three different personas for SELF users, bucketed into the three different exercise categories I discussed above. The personas were based on online research and interviews with my friends about everyday users of tracking apps and wearables. Given the short time frame of this project, the personas were made with a few key assumptions:

  • Users will still be using individual apps to record data day to day, so SELF’s priority must be big picture insights and goals
  • Users understand basic health tracking and are looking for a way to combine their tracking in new ways

I used the personas throughout the project to guide and validate my design decisions and priorities.

Job Stories

I used the Jobs To Be Done and How Might We frameworks to explore a few different use cases for SELF and to empathize with and understand users’ motivations and desired outcomes.

When I train, I want to track all aspects of my workouts, diet, and sleep, so I can perform at my best for my team and be in the best shape of my life.

When summer is approaching, I want to lose a few pounds so I can have a great beach body.

When I had my health scare, I realized that I need to take control of my health and diet so that I can continue to be here for my wife, kids, and grandkids.

Based on the above job stories and my understanding of the types of data available from current wearables and apps, I developed a few user scenarios and tasks:

  • You are an avid athlete with top of the line tracking and wearables, a personal trainer, and a customized diet from a nutritionist. You understand how the different pieces of a healthy lifestyle fit together, but you sign up for SELF for a deeper dive into the data to truly analyze and optimize your health and fitness. (Advanced user)
  • You workout an average number of times per week with a varying mix of cardio, weights, and yoga. You know you feel better on days that you workout, but you’re not sure why. You were just given an Apple Watch for Christmas, and you are excited about all of the new data it will give you. You sign up for SELF to learn how the different pieces fit together. (Intermediate User)
  • You were sedentary up until you had a recent health scare. Your doctor told you you need to make changes and soon, so you went and purchased a Fitbit. You have no idea what any of benchmarks or what the data means, so you sign up for SELF looking for basic recommendations and guidance. (Novice User)

I used all three of the scenarios to guide my process, but I leaned most heavily on #2 as I believe it applies to the largest group of potential, immediate users. One of my goals, however, was to create a product that creates more active people, so hopefully the above assumption doesn’t last long.

Data Sources/Competitive Landscape

I looked through various health tracking apps (both mobile and desktop) to get a better understanding of the data that would be available to SELF. In addition, analyzing the market was important to insure that SELF’s feature set was truly a unique selling proposition.

A quick review also gave me an inventory of existing patterns, strengths and weaknesses to keep in mind as I began my designs.

Identifying User Wants and Needs

I made a list of all of the most common data that is collected by apps and wearables, and started thinking through the most efficient and simple way to correlate and display that data. I used affinity mapping to organize the data into categories, which gave me the basic framework and user flows for the dashboard.

My overarching information architecture for SELF follows a goals structure. I decided to have users create goals, which track progress and create a motivational structure. Data and goals inform trends, which are historical comparisons of your data against each other. Trends are customizable, so a user can compare any and all data points against each other, completely agnostic of source. Trends are the deep dive into the data, the results of which are insights. Insights are smart recommendations from SELF on how to better achieve your goals based on trends.

In addition, the initial MVP’s focus on core features would allow for rapid feedback and iteration on the production app. A goal here was to create clear decisions to validate the design with potential key stakeholders.

The overall theme of SELF is to create motivation through actionable recommendations. Goals, trends and insights combine to create a full picture image of your health that keeps you engaged through active participation.

I kept Nir Eyal’s Hooked methodology in mind when designing SELF, as it states that in order for habits to be formed, the user needs to be sent a trigger that then leads to an action (the habit we want to form, in this case overall health improvement). When the user takes action, a reward encourages them to repeat the action in the future.

In this instance, insights are the trigger and visible goal tracking and encouragement is the reward. In addition, SELF takes advantage of its data sources’ native notifications. Apps such as Fitbit, Nike and Strava already encourage users to get up and moving, so SELF can piggyback off of those notifications through its seamless integration with those apps.

By repeating the trends and insight actions multiple times, the user starts to invest in SELF.

Getting users through the Hook cycles again and again helps users form habits and increases their long term engagement with SELF. Targeted triggers then prompt more action, and through repeated usage, users associate a positive emotional response with SELF. That response turns into an internal trigger that prompts future interaction and engagement with the app.

Nir Eyal also postulates that there are three types of rewards that drive us: rewards of the tribe, hunt, and self.

SELF focuses on rewards of the hunt and self in this iteration, with rewards of the tribe to come as a V2 feature. In this case, a reward of the hunt would be the chase to meet goals and earn recognition in the app in form of praise and badges. A reward of the self is the feeling of competence and excellence from your activities, and a reward of the tribe would be recognition in social competitions and on leaderboards. Rewards of the tribe are also a great way to foster an active and thriving community around SELF, but I left them out due a prioritization of rapid iteration over initial features.

Ultimately, my design decisions were guided by two key user questions:

How are things going today/this week/this month?

What should I be focused on today/this week/this month?

Initial user flow

Ideating the Solution

With personas and flows in hand, it was time to start sketching. I came up with some quick storyboards and sketches for a new user getting up and running.

I did some early validation of my lo-fi sketches and used the feedback to pinpoint my solutions for the hi-fi mockups and prototype.

Prototyping and Validation

The clickable prototype.

I opened up Sketch to create hi-fi mockups and used Marvel to create a clickable prototype. I tested the prototype with a few friends and gym buddies, and did a few quick iterations based on the feedback. Initial feedback indicated that I had too much complexity, and needed to narrow down my initial feature set. Additionally, my first designs lacked a strong hierarchy and information architecture, so it was easy for users to get lost within the app.

I used cards as design elements throughout the app due to their modularity. Using cards allows for easy formatting and customization of data regardless of the data’s original formatting, which was a high priority considering the various potential data sources. The side navigation is also helpful, as it can grow with the app as more features become available. I used cards and a sidenav to create a unified visual system and hierarchy, which was especially important considering the potential complexity of the data sets. I also went with a Bootstrap style 12-column layout to aid with formatting and potential responsiveness down the line.

I utilized a system of progressive disclosure throughout the setup guide to further simplify the process, as users could easily experience cognitive overload if given access to every feature right away.

Forms were kept as simple as possible, and only reveal extra options if the user needs it. I also tried to keep iconography simple and to a minimum, as it is easy to become overwhelmed by icons with unclear meaning.

I created the initial wearable and app setup screen to be very similar to existing app stores, as I wanted the initial cognitive barrier to entry to be low, especially considering that users could potentially have dozens of different app and wearable accounts to connect.

After connecting data sources, users are encouraged to create goals. Goals are completely customizable and measurable. SELF detects all data sources from connected apps and wearables and presents them as potential info for every goal. Users tend to value a product that they’ve spent effort and time on more than a product in which they put no labor into, so goals are a good way to initially create user investment.

Upon completing their goals, users get rewarded with tips and encouragement from Trends and Insights.

Each goal creates trackable trends that offer deep dives into the data. Trends are sortable by category, break down over time and can be combined to offer deep, multi-axis graphs and visualizations. I only show a few potential trends in these mocks, but the number allowed is endless.

Insights are the culmination of trends and goals, and are the main value proposition of SELF. Insights are algorithmically determined recommendations for positive lifestyle changes, based on the relationship between your goals and trends.

SELF aims to turn the quantified self process into a clear and concise funnel, wherein large data sets are unified and distilled down to actionable and motivational tasks for the user.

Insights use dialogues, which is natural language to keep users comfortable while providing a simple interface to understand complex relationships and concepts. I opted for natural language suggestions to keep insights focused, as I felt that there needed to be a clear distinction between the high-level insights and low-level, data-dense trends. Dialogues are a form of reward for the user, as they provide positive reinforcement and encouragement.

Takeaways

SELF is packed with powerful tools to offer a comprehensive, big-picture analysis of users’ lives. My suggested design solutions aim to turn the quantified self process into a clear and concise funnel, wherein large data sets are unified and distilled down to actionable and motivational tasks for the user.

The overall aim of the Goals, Trends, and Insights framework is to create a positive mental model of the quantified self, where users begin to understand and value the relationship between all aspects of their lives. In creating a positive mental model of the easy-to-access power of data, SELF provides lifetime value to the user. By creating a visually uncluttered interface for the aggregation of personal data, SELF helps it’s users understand their success and create and maintain healthier habits and lives.

Given the time constraints, SELF is built around the core data analysis features. Future versions/if I had more time would include social comparisons between friends to motivate and gamify the experience as well as an enterprise level to give coaches a comprehensive snapshot of their entire team. Race and training plans would be integrated as well, so that athletes could best optimize training strategy. I also think it would be very interesting to explore the relationship between people’s medications and exercise, with variables such as time of day taken and activity energy levels.

Gamification techniques such as achievements and leveling up would be used as rewards of the self, as they satisfy our intrinsic need for personal excellence and a sense of competence.

Overall, this was a fun project with a lot of interesting variables. I based this exercise off of existing user behaviors, but it will be very interesting to see how future apps like SELF change those behavior patterns.

Appendix

Tools Used

Sketch, Marvel, Draw.io, Pencil/paper, Whiteboarding

Initial Research Sources

The Personal Analytics of My Life

Fitness Tech and Quantified Health

Quantified Self

I'm passionate about creating great products and improving people's lives through technology.