luckyfrog99’s final look back on a 10-week series of rapid-fire design sprints. Song pairings included.
[Read time: 10 min]
Introduction
In just 10 short weeks, we conducted test after test, combined research across disciplines, and negotiated group product design choices. Although incredibly intense, I’m happy to have had this experience. I have three main reflections that seem interesting to examine:
- The emotional whiplash of optimism.
- A harsh analysis of our design process, based on HCI research.
- Ethics!
Optimism to Pessimism to Optimism
Motion Sickness by Phoebe Bridgers
CS 247B was not just a product cycle. I found myself also cycling between feeling motivated by the beautiful insights of users, and not being able to execute on lofty promises.
Beginning: Optimism
Pride is the Devil by J. Cole
I underestimated the task of behavior change. In the beginning I thought, I’ve seen enough bad implementations and I’m pretty good at behavior change in my life. I drew the Connection Circle and Fishbone models easily, picking out patterned behavior and keystone habits.
Early Middle: Extreme Optimism
6’s to 9’s by Big Wild
We found incredible insights during synthesis. Interviews were so motivating, to hear how users situate their own life experiences in the theory they learn in classes. This is always my favorite part of the product cycle, and I was motivated to create something that genuinely addressed user needs.
Informed by literature, we ruthlessly problematized past solutions, and the criticisms were so exacting that I started to feel like we were on track to create the next biggest thing in behavior change.
Middle: Plummeting Optimism
I DON’T LOVE YOU ANYMORE by Tyler, the Creator
When we began ideating solutions, it hit home how much intention can be thwarted by skill level, and how product design is as much an art as it is a science. With art, you can consume beautiful things and know how to critically examine them, but still not be able to create them. With product design, you can appreciate thoughtful innovations, but not know how to pioneer them.
Even our furthest-reach solutions were only marginally better than existing ones, and though all design choices were backed by previous learnings, I can’t honestly say that our product would break through the common “cold start” problem.
Late Middle: Pessimism
Hope is a Heartache by LÉON
Although we thoroughly tested each stage, I began to lose faith in the product. We implemented good-practice UI standards, incorporated user feedback, and made thoughtful choices. But the app became sprawling with all the added features, and I felt like we could benefit from backtracking, instead of pushing forward, but it was too late in the sprint.
Then, in the swamp of documentation, the tiny nuggets of wisdom we’d gained at each stage started to disappear. (I mean, tell me you would remember every one of your post-its.) We had ninety-nine problems, and institutional memory was one. Even more, we recognized that science, market, and empirical findings are often at odds, and within the team, we sometimes struggled to navigate diverse conceptions of what felt “fresh,” “appropriate,” and “impactful.”
End: Renewed Optimism
Alkaline by Kota the Friend
At our Project Expo, we reflected on our class. Some great implementations, some struggling. We agreed, nonjudgmentally, that only 1-2 of them could actually be picked up. (Ours was not one).
But one note of optimism came back: We and others had given 110% to the design process, and stayed true to good-practice, so at least we’d gotten one bad idea out, and a lot of good experience in. The product wasn’t groundbreaking, but it was a practical 10-week sprint with ample documentation.
Pairing with 347: Rapid-fire attack on the integrity of HCI classes
Song 32 by Noname
At the same time that I was in 247B, I was in the HCI research class, 347, with our TAs Krishnan and Emily, and I learned about the other side of HCI: academia. From that perspective, if 347 is Dr. Jekyll, 247 is Dr. Hyde. Below is the tiniest slice of examples from when 347 problematized 247, and put into question the very integrity of Product@Stanford.
1. User involvement and novel solutions
In 347, we read participatory design literature that showed how designers push too hard for novel ideas and overlook the actual needs and authority of communities. By involving users in more stages of the product cycle, designers can make more informed decisions, but a half-baked attempt at empathy doesn’t mean they create designs that are needed. This is so at odds with the “shiny object syndrome” and “appification” of Silicon Valley that it was hard not to be disillusioned with our own design project.
2. Pluralist feminist lens and an unassuming “user”
While we were creating proto-personas in 247, we were reading about the dangers of implicit biases of designing for nameless “users” in 347. I winced every time I called a potential customer a “user” in blog posts (see this one). It does evoke implicit assumptions, and most people who read about “users” don’t even know that there’s normative biases at play. Then I, now aware of the biases, still didn’t come up with a better term.
3. Thoughtful design for AI
This was perhaps the most painstaking example. As I commented on problematic mixed-initiative AI and criticisms on lazy AI integration for 347, we made a team decision to Wizard-of-Oz a magical AI agent, represented by a cartoon ghost named Boo. I can’t deny the irony.
Ethics: All about capitalism
Milionària / Dio$ No$ Libre Del Dinero by Rosalia
Finally, I’ll leave the class at an impasse with ethics. In response to these criticisms and more, we had little recourse. Christina was an excellent instructor, who posed thoughtful questions throughout the quarter. However, we asked discussed difficult topics like:
What arguments can be made to convince people that centering social good causes can not only improve a company’s morality, but advance their “bottom line” offerings? How would you recommend these responsibilities be divided, and what kind of oversight should there be?
And I found my response more often gravitated towards:
Well, in a different economic system with non-monetary motivations … But the best we can do now is …
It was easy for me to identify problems and blame systems — the world’s vices are largely fueled by institutionalized laziness and greed — but it’s much harder to pinpoint what my own perfect system of ethical practices would look like, much less how to position a designer in this mix. While my analysis halts here for now, this class was an early attempt at praxis, at least.
Final thoughts
La Vie – Ichon
I wrestled with a lot of questions in this class. And I’m not even close to nuanced answers. But, at least with 10 weeks of documented work, I have a few solid takeaways:
1. Documentation
This one, is simple. Documentation is important, and it’s a job of its own. It’s important not only for your own sanity later down the line, but for showing others your product evolution.
2. Creative direction
This one, I’m intrigued by. A major criticism of our final project was how piecemeal it turned out to be. The most successful UI in our Project Expo, and the best products I can think of, all have clear creative vision. Our team handled diversity of thought in a tepid way. Thus, compromise between team members sometimes came at the price of low creative direction. While I don’t think this was a bad approach, I’m interested in centralizing creative decisions in future projects.
3. Product cycle
This one, I’m most sure of. I’m learning that the product cycle is a design template. It’s a brute force, quantity-over-quality tool, that despite your most thoughtful efforts will sometimes lead to a bad product. I’m grateful that we got to practice thorough execution and documentation at each stage, under great mentorship, but as I go on to many more design processes in my future, I will continue to refine my critiques into actionable insights. I guess it’s work cut out for a product designer.
