Designing for Complex Values
Updated: Jan 12
In my last blog, I talked about how inaccurate clarity and over-simplification of complex values can make users’ lives worse. As we saw with MyFitnessPal, users can download an app with the aim of improving their health or fitness and instead get sucked into caring about eating under a certain number of calories or walking 10,000 steps a day.
In this blog, I’ll address how UX designers and writers can better communicate complex values to improve users’ lives. I’ll present a redesign of MyFitnessPal to illustrate a better approach. While there are key differences that separate health from other complex, value-laden endeavors such as language learning or building a community, there are a few common lessons that emerge from this exercise.
Designs that deal with complex values create problems for users when:
The complex value is simplified to one or two numerical metrics, e.g., health = eating under 1620 calories + walking 10,000 steps a day. It’s easy for users to get obsessed with these numbers and try to achieve them, even if it’s at the expense of their overall health. I know I’ve walked those extra steps to meet that number even when my knee was bothering me and I needed rest.
The complex value is simplified to only one of its component parts, e.g., health = getting a good night’s sleep. There’s a lot more that goes into being healthy than getting a good night’s sleep. There’s mental health, nutrition, movement, and living in an environment that hasn’t been neglected or purposefully polluted. We are often, easily misled into thinking one aspect of health will fix everything.
Users have no control over which aspects of the complex value are emphasized, e.g., the user can’t remove their step counter from the MyFitnessPal homepage, even though they are a wheelchair user. If a user wants to track whether they are meeting 10,000 steps a day, that’s their prerogative. It shouldn’t be decided for them.
(Things get more complicated when we start to look at community standards and who should decide, for example, which rules of social conduct should be enforced on reddit.com. Even at that level, user control and input matters, but the negotiations at the community level look very different than an individual interacting with a personal app. Let’s limit things to the individual level for now.)
These problems are not exhaustive, but they provide a few central guidelines for designing for complex values:
1. Don’t assume the user’s values. Let the user choose which aspects of the complex value to focus on.
2. Make sure the app has enough complexity to communicate a variety of key aspects of the complex value.
3. If you use any numerical metrics, explain how the numbers are calculated, allow the user to customize them, and de-emphasize the numbers on the app to avoid user obsession.
Case Study: from MyFitnessPal to MyHealthCircle
Let’s try to take these lessons and redesign a health and fitness app. Call it “MyHealthCircle.” The basic idea of this app is that users can select, customize, and adjust their health goals to track them weekly through a color-coded wheel, where each element fills up or decreases based on how well the user has met their goals over the last week (this will be clearer in the wireframes below).
The way I've conceived it, the circle/wheel and any reminders or popups would encourage users to see the app as a kind of dashboard to prompt users to look under the hood of their health. The “engine light” only gives you so much information, but it can still be a useful tool, especially for those who don’t have a good intuitive sense of how they’re doing.
In what follows, I’ll first present the problems I found in select MyFitnessPal screens and then explain how my redesign solves those problems. (My redesign is not complete, but it showcases key features and differences.)
Here are the first two pages the user encounters when they download the MyFitnessPal app:
At first glance, there seems to be a variety of goals and options available to the user. However:
❌ While there are seven options for the user to choose from, they fall into just three categories: weight management, exercise, and stress management.
❌ There is no option to set custom goals.
Here are the first two pages I’ve created to introduce the user to MyHealthCircle:
In this redesign:
✅ The areas the app focuses on are much broader than those presented by MyFitnessPal.
✅ The user can add their own custom goals, trackers, or reminders.
✅ By placing each of the user’s chosen categories as equal pie slices in a circle, no one category is automatically prioritized over another.
Next, let’s take a look at select screens from the MyFitnessPal setup pages:
As I discussed in the last blog, the main problems are:
❌ The app still gave me a target calorie limit, even though I selected the option to get strong without worrying about body fat or definition.
❌ There is no explanation of how exactly they calculated my calorie number.
❌ There is no acknowledgement of intersex people (even in the linked article) nor of disabled people whose calorie needs may require adjustment because, say, they have had both legs amputated.
Here’s a redesign for the nutrition setup for the MyHealthCircle app:
In this redesign:
✅ The user is only given calorie goals if they selected that option.
✅ The user has the option to customize their goals and ways of tracking their nutrition at multiple points in the process.
✅ The user can access an explanation as to how their calorie goal is calculated, and there’s an acknowledgment that the calorie calculation process may have missed key information.
✅ The biological sex slider recognizes the existence of intersex people.*
✅ In both the first and last screens in this process, there’s a built-in reminder that nutrition is more than just a calorie count.
✅ I’ve built in a calorie range instead of a single, daily number, and I’ve set up the tracker to look at average calorie intake over a week. These changes should create some flexibility and discourage obsession with the daily number.
*I’m not sure the slider is the best way to communicate this, in part because a condition like PCOS which raises androgen levels in women tends to lead to weight gain rather than to weight loss. Additionally, the slider may be confusing to those who fit neatly into the biological sex categories of male or female. My solution is just one possible redesign.
Finally, let’s look at the home screen for MyFitnessPal:
On this home screen:
❌ The first text and trackers the user is greeted with encourage daily attention to a single number, which can lead to obsession.
❌ The user is automatically given a calorie, steps, and weight tracker, whether or not they expressed interest in any of these metrics.
❌ While users can customize their goals for each of these trackers, they can’t remove them from the page (at least as far as I was able to figure out when toying with the app).
Take a look at the new MyHealthCircle home screen (pretend the three elements in the circle are clearly labeled and separated by more than just color):
In this redesign:
✅ The only trackers that show up in the circle are ones that the user has chosen and had the option to customize.
✅ Likewise, the resources the pop up on the lower half of the app are tailored to the goals the user has chosen.
✅ I’ve made it very easy to adjust the contents of the circle.
✅ There are no numbers on the home page, to avoid oversimplification and discourage obsession.
✅ At the same time, users can log their information daily and analyze long-term patterns.
With this redesign, users can access the same functionality as other fitness apps, but the app’s organization promotes a focus on maintenance rather than linear progress.
There’s a lot more to work out than I’ve had the space and time to do here. And there are likely still opportunities for oversimplification and obsession, perhaps now with the gauges of the weekly circle. I’ve tried my best to avoid those tendencies by creating a more holistic set of goals and enhancing user autonomy in the selection of those goals.
As I said last week, I’m still not sure that there’s a way to completely fix MyFitnessPal. This is my best initial attempt at that project. Given the tendency of users to get obsessed with clear metrics, there might not be a way to create a clear, gamified health app without encouraging user obsession, especially given the background culture we have around weight and exercise. At the end of the day, however, I’d much rather err on the side of autonomy than paternalism.
This app redesign is meant to encourage user autonomy, by providing users with a more accurate conceptual framework and the ability to choose their own goals. If there is a guiding ideal for this redesign, it’s informed consent. I’m not sure if the redesign has totally achieved that ideal, but it’s certainly a lot closer.
Photo Credit: Gwyn Vandergaast