Final Reflection – Isabelle Zhou

Introduction

Before this class, I thought that designing for behavior change simply meant creating affordances and nudges to point users in the right direction. I did not know much about the nuances of the field, much less the importance of understanding the user and the bigger picture of ethical considerations.

Experience

Our team worked on Remindful, an app that helps college students living away from home keep in touch with their family. It uses notification nudges and a friendly progress tracker interface to target college students who already have good relationships with family (high motivation and ability to keep in touch) but may forget to do so or feel that they don’t have the time to contact family members regularly. 

I loved how this quarter focused on understanding the user, and running longitudinal studies in the user’s natural habitat to get to know their habits and behaviors. This approach makes a lot more sense to me than running quick sporadic experiments (i.e. 30 minute interviews) that are assigned in other design classes. One downside was the short 10 week nature of the quarter, so our conclusions and extrapolations all took place in a limited time period – there were several other avenues I hoped to explore around family relationships (i.e. do behaviors change around holiday time) but simply did not have the time to do so thoroughly. 

A specific problem we encountered with our project occurred during the design phase. When choosing the visuals and color palette for our final app, we had a discussion about whether our colors and fonts were too “cutesy” and not appealing to a broader audience – we ended up concluding that the warm colors and cartoon flowers would appeal to most college students. However, after a closer look at our moodboard and why we chose the colors we did, we decided to lean into the “cutesy” design rather than changing it to a more minimalist design because we wanted our app to be extremely loved by our target user (who would enjoy the warm feeling and design of the app) rather than somewhat useful to a broad swath of users. 

On a personal level, I appreciated how this class humanized the end user and contextualized behavior design with ethics discussions. As someone currently working in early stage venture capital, I see far too many fledgling companies design products that might be lucrative but have incredibly negative second and third order effects. This class has taught me to use my position of power to speak up during these early startup meetings and suggest ethical perspectives to founders before their company grows bigger and impacts scores of real users. 

Ethical Considerations

Our app, Remindful, used 2 mechanisms to change behavior: 1) notifications to remind users to contact their family and 2) home page using plants to depict the  status of last contact for each relationship. Mechanism 1 is an ethically acceptable nudge because we only notify users during a time frame that THEY choose. To sign up, users have to input what time of day they have downtime and want to contact their family member. We are helping users accomplish something they already want to accomplish with a quick notification. Mechanism 2 is trickier but still an ethically acceptable nudge because of user safety features we built. For example, we understand how using a virtual plant to represent the health of a relationship might be harmful because it turns an internal motivator into an external motivator. To prevent users from gamifying the system (i.e. calling their mom 20 times in a day to maximize plant growth), we set the plant to only grow at certain rates – i.e. once a week if the user calls their family member – so that we can help users build a good habit over time. On the flip side, the plant doesn’t immediately wilt or die if a user forgets to call their family member one week (since we know things can happen). We merely give the user a notification and then a second chance – if the user continuously forgets to call their family member, then the plant begins to wilt. 

Because our target audience is fairly niche, we looked to design our app to be as universal as possible for the majority of our target audience. We minimize scrolling as well as the number of pages on our app for broad ease of use. Our app also does consider a few edge cases (not fully implemented yet), notably a dark mode for visually impaired or light-sensitive users and a voiceover narration for blind users. The tradeoff we made with our home page was to use an easy-to-understand plant format for the majority of our users, or to use a harder to understand list format that is marginally easier for visually impaired users. Considering how the plant format has significant benefit for the majority of users, and the list format only has marginal benefit for our edge cases, we chose to go with the plant format but with elements of voiceover narration on the future roadmap.

Conclusion

Ten weeks later, I’ve realized that behavior design is so much more, and requires a deep understanding of users as well a careful consideration of the overarching impacts of technology in their lives. Next time when faced with a similar situation of designing for behavior change, I will first and foremost seek to deeply understand my target user so I can create a product that actually resonates with them rather than something I think will resonate. Going forward, I’ll also take with me my learnings around ethical considerations into the real world, whether it’s to launch my next startup or to support early stage founders (aka the large tech companies of tomorrow) in the right direction of building products that are net good for the world, rather than net negative.