Before this class, I thought that designing for behavior change was honestly a useless endeavor. I did not see the value in trying to force people to change their behaviors because I was of the belief that true change has to come from within. I believed that this was really all just a matter of motivation and willpower. While I do still believe in this to an extent, this class has opened my eyes to the power of effective design for behavior change. Perhaps my initial naivety stemmed from my own difficulties with behavior change, so I do appreciate how much this class has taught me about how we can design products to help people form better habits and decrease bad habits. For our group, Brevi came from our personal struggles with getting good and effective work done while trying to avoid burnout. Like so many of our peers, our default break activity would be to scroll on our phones. While we have talked ad nauseum about how difficult it is to truly address doomscrolling as a habit, I hope that Brevi could at least inspire some positive change within the scope of taking breaks in between work sessions.
One of the things I loved most about this class was working with my group. Although we had met for the first time in the class, we each understood the role we had to play and the amount of time and effort we would need to dedicate in order to create a successful project. As we divided the work amongst ourselves, it was amazing to see the progress we made over the course of the quarter. While I have worked on many group projects in the past, I can confidently say that Brevi as it is today would not exist without the help of my wonderful group. And if I had to choose one thing that I hate about working on the project over the past ten weeks, it would be that it is now over.
While I have taken plenty of HCI classes in the past, this is the first time I am taking one after finishing CS 147. A lot of lessons that I learned in that class, from good UI/UX to the importance of user testing and acquiring feedback, found their way into the approach our team took as we worked on this new project together. In particular, the rapid prototyping and iteration that has only recently been made possible thanks to vibecoding and agentic coding allowed our group to make something we are really proud of in just the final few weeks of this quarter.
That being said, I think that vibecoding in its current state is still severely lacking in the design department, which is why I was happy to be able to contribute my design expertise to help make our app look great. Additionally, I think the exercises with the moodboards and style tile were quite enjoyable and useful, as the final style tile we created for our project ended up being something we could pass along to the coding agent to swiftly implement throughout our app. In the past, this is something that would have taken weeks to overhaul the visual design of an app. Thanks to Claude Code, it took less than a day. Wow.
If there was any major problem to point out while working on this project, it would certainly be a matter of how quickly the quarter went. Creating a fully working app from start to finish was a large undertaking, even with the power of Claude Code by our side. Given more time, I can certainly think of ways that we could further improve on Brevi, perhaps even bringing it to production as a real mobile app. Had we had more time and more bandwidth over the quarter, perhaps this is something that we could have achieved. I think that one large remaining gap for the final prototype we made is that it is a desktop-first app. This is largely a result of focusing our initial development efforts on Bolt, which gave us a non-React Native project. But by the time we ran out of Bolt credits, we already had a somewhat decently working desktop app, so we decided to continue forward in that direction. I think that given the opportunity, we would have wanted to redesign this app to be mobile-first. Since most users are expected to be doing work on the computer, delegating the app to their phone screen would both allow for viewing the timers and such while working and also prevent the user from using their phone for other non-essential tasks during a work session. Perhaps another time.
Nonetheless, I think I will be able to look back on this class ten years from now quite fondly. I am sure the landscape of software development will look radically different by then. Perhaps by that point, Claude will even be doing all the prompting for me. But I will at least remember the time I spent working on this neat little app with my groupmates.
Ultimately, our app relies on nudges in the form of giving the user cues of when to work and when to take breaks through visual indications on the screen as well as audio pings to alert them. I would argue that these are acceptable nudges as they are largely just suggestions for what the user should be doing. After all, Brevi does not lock the user out of accessing other apps on their device, and there is no true way to determine whether or not a user is following the guided activities. But I think that if a user were to go through the trouble of setting up their work sessions in Brevi but not follow them, then they are really just wasting their time. Ideally, the app does not stray into true manipulation as the guided activities and work session paces are merely meant to be a guide that the user can choose to follow. In fact, our app supports the user’s autonomy to extend their current focus session or change what break activity they do. Our goal is to supplement a user’s work session, not micromanage everything they do.
Additionally, our app is great for user privacy as we do not collect any sensitive information about the user. Additionally, the only persistent data that is created, regarding work sessions and user preferences, are all local-first and get anonymized whenever saved to their account in the database. One way to improve our commitment to privacy even further would be to keep everything local, but this would be inconvenient for users who may want to use the app across multiple devices. At least for the time being, Brevi collects as little information as possible on the user to allow for cross-device support.
Our usability testing also revealed a lot to us in terms of interface design and design justice. To start, we did not take into consideration users with differing abilities when designing break activity types. This means that a physical activity like walking could be suggested to users who may be unable to walk. We sought to improve this by allowing users to select to opt-in and opt-out of different activity types during onboarding, without making direct mention to single out users with differing abilities. This was a case of making a universal design that was able to address all users equally. Overall, we wanted the design of our app to be done in such a way that just about anyone could use our app without feeling like they were being singled out as a special group for one reason or another.
After taking this class, I now think that designing for behavioral change is not only possible but also important. In many cases, users may want to change their behaviors and have the motivation to do so, but frictions that arise in their lives make commitment and consistency difficult. But by designing interventions that are designed to help change behaviors in one way or another, these products can help people to love the lives they want to live. I think that is particularly powerful, and it is certainly something I will keep in mind as I work on future projects.
Next time when faced with a similar situation, I will make sure to take a step back to consider all the stakeholders involved, especially the users who will be using the app. I have recently gotten into the hobby of creating my own apps, and I think many of the lessons I have learned in this class will be immensely useful. One thing that I really like about all the HCI classes here at Stanford is that they all complement and build upon one another. By the time I graduate, I hope I can learn the skills necessary to make really cool designs that can help people in one way or another.
Thank you for a great quarter!
