Before this class, I thought this…
I came in thinking design was mainly about making something intuitive, then validating it with a few user tests. I also assumed ethics would be mostly a checklist item, like privacy language and avoiding obvious harm. I did not realize how much the framing, defaults, and tone of an interface can be the intervention.
I did this work with these experiences…
Our team built a pauses app that adds a little friction when someone starts a break and suggests a more restorative option. We did interviews, built personas, used diary-style logging formats, made a moodboard and style tile, and iterated through onboarding, break flow, and review. Usability testing helped us identify what blocked the core loop and what made the app feel judgmental or inaccessible.
What I loved and what I struggled with
I loved how the course made small behaviors feel meaningful. Pauses sounded trivial until we heard people describe doomscrolling as anxious anesthesia, or lab work as time-blind and perceived, or movement as the only thing that resets them. I struggled with the ambiguity of building a well-being tool in a productivity-shaped world. It was easy to drift into optimization language, and we had to keep asking what we were rewarding and what we were accidentally reinforcing.
What worked and what did not
The research-to-prototype-to-test loop worked. It prevented us from building based on vibes. The hard part was that habit tools show their value over time, and a prototype test can only capture a slice. That made me more careful about what our tests could truly claim.
Methods I will reuse
I will reuse interviews, lightweight diary methods, and assumption tests. I will also treat microcopy and defaults as first-class design work, not polish. Small choices like asking What did you need? instead of What did you do? can change honesty and shame.
A specific problem and what we learned
A major issue we hit was inclusivity. Early break suggestions assumed walking, going outside, and having energy. Testing made it clear this excludes people and contexts. We added onboarding preferences to filter suggestions by ability and environment. That taught me inclusion is not an add-on, it changes the product structure.
Ethics, more personally
Our app uses nudges, and they feel acceptable only if users opt in, maintain control, and the tone stays supportive. The same mechanism can become manipulative if it uses guilt, pushes one definition of a good break, or escalates reminders. Privacy matters because break logs can reveal stress and avoidance. If this data ever becomes something an employer expects, the app shifts from support to surveillance, so strong privacy defaults and clear boundaries are essential.
Now I think this…
Now I see design as values work plus systems work. We cannot treat behavior change as neutral, because the interface shapes what users feel about themselves. I also think the most important part of well-being design is planning for low-capacity moments, not ideal behavior.
Next time, I will…
Next time I will name value tradeoffs earlier, test framing and tone as seriously as navigation, and design for the worst day. Ten years from now, I think I will remember that we were trying to build a calm place inside a loud world, and that kindness in design is not just aesthetic, it is a safety feature.
