UpBeing is powered by advanced machine learning as well as emotion detection and recognition technology. But the app’s most important interaction has nothing to do with algorithms or code. Check-ins are the feature that connects UpBeing’s passive data collection and machine learning with the feelings and experiences of real people. Without it, nothing works.
Below is a conversation between Andrew Fullerton, Growth Coordinator at UpBeing, and Amanda Connolly, UpBeing’s Lead Product Designer, discussing the evolution of UpBeing’s Check-In feature.
Andrew: When I use the UpBeing app today, the importance of check-ins seems obvious. But when I think about it more, I realize that there were so many other ways you could have chosen to collect input from users about their feelings. So, why did UpBeing need a Check-In feature at the centre of the app?
Amanda: I really think of check-ins as the keystone of what we’re doing at UpBeing, and that’s why it has so much prominence in the app.
Collecting passive data and linking your devices and calendar is relatively straightforward – at least from a user’s perspective. It enables the app to analyze how you spend your time. However, what I think sets our approach apart is our emphasis on understanding (and helping users understand) your emotional state during these activities and events in your life.
Some people are super productive when they have a packed schedule; they’re able to thrive when they’re really busy. Some people feel overwhelmed under the same circumstances. Check-ins are the way we make the connection between what’s happening in your life and how you’re feeling about it.
The other important piece of these check-ins is how habitual they are. Building emotional intelligence is a huge part of being able to improve your wellbeing, but being able to uncover hidden patterns and insights in your life is an equally important piece of that. Journaling takes time. Reflections are often too far removed from the events of your day or week to really capture how you were feeling in the moment. The check-in is really meant to be quick and built into your day. It’s the gateway to exploring the intricacies of your life.
Andrew: What did check-ins look like early on and how did they get to the point they’re at now?
Amanda: A year ago, UpBeing was a desktop app with a data-heavy dashboard. In my first week after joining, we started our first round of user interviews.
The feedback we received was essentially:
I knew I needed to understand our users, and the problems they were facing, more clearly. I had sat in the interviews, and I had heard all the feedback. But I was stuck – did we simply add some new widgets to the dashboard and time-release the check-ins? In my gut, I knew that wasn’t the right direction to go, we needed something more.
So, I used this concept called noun foraging. Usually, product design is centred around the things you can do (verbs) in the app – Jobs to Be Done is a good example of this orientation. What noun foraging ultimately reveals are the things (nouns) in the user’s mental model that they are using to understand either your application or the problem space you’re in.
So I took all the user interviews, transcribed them, extracted all the nouns, and then ran them through a crazy spreadsheet that identifies and counts the nouns in common across all our user interviews. The number one word across all our interviews was: “day”. There were over a hundred instances of people using the word “day”, even though the concept of a single day (at that point) didn’t exist in the UpBeing app. This gave me the exact clarity I needed.
From there, everything changed. Structuring the app around concepts of one-day-at-a-time, forged a new information architecture, created more clarity around check-ins, removed the dashboard view entirely, and took UpBeing from desktop to mobile.
Andrew: Alright, so Day View is in, and the dashboard is out – tell me more about that.
I wanted there to be a really strong narrative between completing check-ins and having the app reflect that input back to users – without a dashboard. Our user interviews told us that people have a hard time thinking about their wellbeing in numerical terms. And for me, numbers ultimately lead to thoughts of – “Am I enough? Am I not enough?” and I did not want that for our users.
Often I see people, places, and memories as colours – I have memories of days that were so joyous I think of them as bright pink and yellow. Or people, who – without any further explanation – I see as swirling shades of grey and black. So I played with this idea, “What would it look like if we reflected days and feelings back to users as colours – instead of numbers?” When I presented my initial sketches back to the team, I was a little nervous – would they think I was weird? But it was met with such enthusiasm, I knew we were on to something.
So I introduced this concept of figments (or blobs as we lovingly refer to them) and the orb. Figments are shaped by what you’re feeling, coloured by how motivated you are, and scaled by how productive you’ve been. Initially, these figments had been designed as part of our external brand, but they found a new home, and new importance in the app. The orb, I saw as a container, and also as a mirror. It was a reflection of everything a user was putting into the app, and a way to look and identify with how you’re feeling.
From there, it was easy to create the thread of the figment and the orb through the user experience. Each check-in a user completes throughout the day generates a new figment that is housed in an orb. And I felt like that really reflected real life, the way we move through the day, feeling different things as life happens for us.
It’s a way of communicating: “Hey, this feeling is just for right now. It isn’t going to be forever and maybe it’s not even going to last for all of today.”
Andrew: That is so true. One of the amazing things about really great UX/UI design is that it makes interacting with apps feel frictionless. But there’s a ton of thought, research, and hard work that goes into making these things feel seamless. What were some considerations you kept in mind when designing the check-in experience?
Amanda: One of the questions we asked our early users was “What is the most time that you would willingly spend on doing a check-in?”
Most people said one minute or less, everyone felt more than 5 minutes would be a no-go.
So now we had a time constraint. But there was another challenge with it. We didn’t just want UpBeing to be an app that tracks emotions. We wanted UpBeing to help users understand the relationship between what they do and how they feel – to help our users understand why they feel the way they do.
Finding a way to collect emotional input from users in a minute or less is challenging. Finding a way to collect emotional input from users while helping them think deeply and differently about their emotions in one minute is a much taller order.
We explored quite a few different models for doing this. We considered Likert scale survey questions, mood wheels, and even an emoji scale. The Likert scale questions ultimately were just too tedious for a check-in process that needed to be completed multiple times a day. The emoji scale was interesting, but I just felt like feelings are a bit more nuanced than “Smiley-Face” vs. “Angry-Face”. The mood wheel was a really strong contender. I actually still really like the mental model of there being primary emotions the same way there are primary colours, and then more complicated secondary and tertiary emotions.
Ultimately, though, we decided to model our check-in process on a version of the valence-arousal scale.
The way that this particular chart provided a way of mapping emotions along two dimensions with valence (simplified to attitude in the UpBeing app) running along the x-axis and arousal (simplified to energy in the UpBeing app) running along the y-axis, gave us a meaningful way to interpret a user’s emotions without requiring them to answer too many questions.
Andrew: One of the things that strikes me about our check-in process is that we’re essentially collecting data on something that’s completely subjective. Feelings are completely subjective – we have ways of describing them, but there’s no way to be truly objective in labelling them. How did you help users navigate this challenge?
Amanda: Everything about the check-in feature centres on choice. Users need to have enough choices to account for the subjectivity of feelings. You and I could be experiencing the exact same physical symptoms and yet experience our feelings in a very different way. But users also need enough constraints so that they aren’t paralyzed by choice.
Think about what happens when someone asks us how we’re doing. Generally, we say we’re “good”, “fine” or “not bad” followed by a “how are you?”.
The question “How are you?” is so vast and so routine that we defer to the same few emotions we always think of. There isn’t enough specificity for us to have to think about how we’re doing. This, in particular, is one of the reasons I never really explored an open-ended or text-box style check-in. We needed users to actually pause and think, and moreover, we needed to give them the language to be able to describe how they were feeling.
“Language is our portal to meaning-making, connection, healing, learning, and self-awareness. Having access to the right words can open up entire universes. When we don't have the language to talk about what we're experiencing, our ability to make sense of what's happening and share it with others is severely limited”. - Atlas of the Heart, Brene Brown
This is why, when you start a check-in with UpBeing, there are two questions right off the bat.
Initially, this was a single quadrant that asked users to identify what they were feeling as:
But what we found when we tested this with users was that they struggled to separate energy level from attitude, and had no concept of which moods lived in each quadrant. So, we broke up these four options into two simple binary questions: positive or negative, high or low.
These questions serve two purposes in the app. First, they’re a way of priming users to get granular in thinking about their feelings. Secondly, when users answer these two questions, the UpBeing app drops them into a quadrant in the mood grid that correlates to their responses. Once in this quadrant, users can press and hold any feeling to read its definition, to understand if that’s how they’re really feeling.
While we do drop users in a particular quadrant, we don’t suggest any particular mood to our users. This was also an intentional design choice. By showing users a range of moods that match what they might be feeling, users are encouraged to actually tune in to what they are feeling. Sometimes we think “I’m so angry right now!”, but then once we see “frustrated” or “irritated”, we reconsider. “Maybe I’m not angry, maybe I’m just feeling irritated.” There are definitely times when I’ve committed to the anger, but oftentimes I’m able to find something else that aligns better with what I’m feeling.
In earlier versions of the app, we actually locked users into a particular quadrant of the mood grid based on their responses to those two questions. We’ve since changed that. We found that users actually got a lot of value from having the freedom to explore what they’re likely feeling in relation to everything they could be feeling. It was also important to us that if you feel another area of the grid better captures what you’re feeling, you’re free to explore. We really want people to explore their feelings.
Interestingly, one of the things that we’re thinking about right now based on user feedback is how users can choose more than one mood. One of the things that we’ve heard from folks is that sometimes they’re not feeling just one emotion: they’re experiencing a few different ones.
Andrew: Last question. After you select your mood from the grid, there’s a final set of questions about motivation and productivity. Where did those questions stem from?
Amanda: Helping users label their feelings and build emotional intelligence is part of our mission at UpBeing, but it isn’t all of it. Our goal is bigger than that – we want to improve the wellbeing of individuals, communities, and society. Emotional intelligence is part of that mission, but another big piece of that is helping people figure out what in life is meaningful to them.
Initially, we were trending towards a really heavy questionnaire – I think it would have been something like a series of 10 questions on a scale before you got to say how motivated you were. But I knew that we would lose people by doing that, and I was still holding the “one-minute-or-less constraint”.
So what we did was we actually flipped that scaled question approach on its head. We decided to ask users how motivated they are first and then let them add context through a series of “I” statements where users can just check off what’s making them feel either motivated or productive.
In the morning check-in, we ask users about their motivation. How do they feel about what they have to do or should do today? In the wind-down check-in, we ask users about their productivity. How do they feel about what they ended up doing today?
It’s all about asking the right questions at the right times to help our users think critically about their wellbeing and behavioural health in meaningful and different ways.
Amanda is the Lead Product Designer at UpBeing. She joined UpBeing in 2022 and is the first in the role. You can connect with her on LinkedIn.
Amanda, Andrew, & UpBeing