Part 1: Rethinking Step Tracking Visualizations
Demand for personalized health monitoring devices has skyrocketed over the last decade. Hundreds of millions of users monitor health metrics in real-time at a relatively low price point using wearables such a watch or band.
As promising as this space is what I've found to be disappointing is not the hardware or sensors that enable data collection but rather the "cooker-cutter" approach applied to data visualizations featured in health metric dashboards. As an aside, check out Willem's blog to learn more about research into how health is visualized in fitness apps and games.
Take for example the widely adopted health metric for monitoring daily physical activity referred to as step tracking. Each step is calculated using a 3-axis accelerometer to count cumulative steps throughout the day. The end result is a single data point or step count consisting of the total or sum plotted as a bar chart (or similar chart type). As a user the oversimplification bodes well for gamification and engagement and yet omits opportunity to convey process feedback that vital to behavioural goal setting.
What is a Step Narrative?
Stories connect us as humans and are an effective means of learning. Stories present narratives that help us to understand that information and each other, and it makes the information memorable. Therefore, data visualizations that tell the best stories stand to benefit users most.
Let us better understand the concept of a stepping narrative by introducing a definitions and example that builds on itself.
- Step: movement of putting one leg in front of the other in walking or running.
"...[400 steps]...[70 steps]...[1100 steps]...[0 steps]..."
- Narrative: a spoken or written account of connected events; a story.
"In the morning I walked around the mall until I found the store that was selling the perfect outfit, which I purchased on sale. After that, I walked directly to the office arriving on time for work. For the remainder of the work day I sat at my desk before taking a taxi home. The kids stayed at my parents place so my partner and I took easy. We ordered pizza and watched a movie".
- Stepping Narrative:
In the third example step values are woven into the sentence structure. Note, the narrative of the story or daily narrative report remains in tact despite having adding numbers into the mix.
"In the morning I walked around the mall until I found the store that was selling the perfect outfit, which I purchased on sale [2200]. After that, I walked directly to the office arriving on time for work [1200]... we ordered pizza and watched a movie [0]."
The next iteration is to determine what type of graph or chart best lifts the text and the numbers off the page while keeping the narrative in tact. For simplicity sake, the data fields are limited to date-time, and steps per minute. To my best knowledge, the running line chart (as plotted in the stock chart below) is most appropriate for displaying continuous data in time sequence.
Time: The element of time structures each step into a linear progression of events unfolding in sequence. Without time, much context is lost, and the data floats into abstraction - removed from the personal narrative altogether.
Cumulative Sum: The y-axis capturing the magnitude or intensity of the event.
Together time and the cumulative sum of the activity form a distinct shape or pattern that is simple yet rich with context for the users make sense of.
From Numbers to Shapes and Back
Consider the four curves shown below. Each shape is recognizable by the human eye and while you may not be able to put words to it when read from left-to-right the pattern is as intuitive as anything.
I anticipate with practice that users will find the simplicity of cumulative line chart to be sufficient a mental cue as any to provide timely feedback that helps to inform the learning process and future iterations typical of behavioural goal setting.
TOP-RIGHT: steep curve that increases rapidly for 1 minute followed by a subsequent plateau or flat-line indicates patterned physical and possibly event the activity itself and the context surrounding the event
"I was walking to the bus at snails pace when I realized the bus had closed its doors and was now pulling away from the stop. At that point, I had no choice, I sprinted to chase after it. Thank goodness I did, otherwise the bus driver wouldn't have seen me in the side-mirror."
Rethinking Physical Activity Dashboard 2.0
The goal is to design an intuitive, customizable UI featuring cumulative sum of steps + filter capabilities that, when taken together, help fosters a sense of curiosity and empowerment whilst reviewing intraday activity patterns over days, weeks, months and years. Filtering creates custom views that can serve as a window into user preferences for how they prefer to use the UI to set and monitor goals.
The design should also optimize for the amount of actionable information that can be gleaned under the shortest period of time with key product outcome measures as (a) number of daily users, and (b) number of goals set and achieved.
Stepping Narrative Dashboard Wireframe
The UI introduces the following features:
Multiple drop-down fields for users to query as they see fit
"Account Name" and "Date" filters are required whereas other filters and trend indicators are optional
Cumulative steps are plotted by default
Step targets are configured in goal setting window and is plotted by default
A bar chart is positioned adjacent to step count % relative to goal or sub-goals
Plus/minus buttons are available both on the x-axis to allow users to drill-down on specific segments of the timeline
An image of a person is annotated to the bottom left-hand corner of the chart to emphasize the storytelling element
Matplotlib
Next Steps:
In a future blog post 2.0 I will share the steps taken to design and solicit feedback from active users of popular wearable devices (Fitbit, Apple, Garmin, etc) by way of survey data. Responses will inform future wireframes.
Stay tuned!