It started with a vision quest.
As a founding member, this project was basically a blank slate. Early on, our team was a technologist, a business-minded visionary, and me. That meant I wore a lot of hats: Branding. Research. Journey mapping. Learning new UI patterns. Concepting and validating ideas. Pivoting based on investor feedback. Establishing design principles.
The initial prototype was a proof-of-concept that machine learning could be applied to events. We scrubbed data from event sites, then made users choose whether or not they were interested in an event. In this example, we used previous answers to suggest more jazz events.
Briefing and prototyping.
We had users in mind when we started (20 somethings like us), but I'd be lying if we didn't have a bit of Social Network mania going on. We agreed that our time would best be spent making an iPhone app — there was no direct competition at the time, and casual inquiry suggested that people were excited to to be able to search for and share events on the go.
Because some of our team believed machine learning was the real value in our company, my designs had to give the ML scientists the training they needed. The most successful early prototype ended up using a Tinder-style interaction, where you'd swipe left if you're interested in an event and right if you're not. Based on your answer, we'd learn what type of events you most are likely to enjoy and present you with better answers in the future. At some point, investors questioned why people would want to spend such a long time swiping when they could go onto timeout.com and quickly see everything that's happening where and when they want.
We stopped requiring users to give feedback on events and moved toward a list view, hoping that users could find something engaging more quickly. We added a filter system so users could refine the list while we we still learning about them.
Ch-chain-changes!
Even though our proof-of-concept was able to land us some funding, we had a long list of investor feedback to address. The overhaul involved a full redesign of the app, including adding search, social functionality, and ticket buying. I spent a lot of time in the details, focusing on working with engineers and embracing Apple's skeuomorphism. In hindsight, the team actually needed to prioritize their product efforts. This was a first-hand lesson in why design thinking processes benefit the entire team.
When a user touches the filter bar in the list view, they can narrow their suggestions. I used icons to save on space, and added the glow as an affordance so users could track where they touched.
Hindsight leads to insight.
Unfortunately, this story doesn't have a happy ending. But, I still include this piece precisely because it taught me that every discipline benefits from design thinking.
Successes:
I produced an enormous amounts of design assets, collaborated from 0-1 with what would become the holy trinity of product design, and taught a junior designer the basics of digital design. I adopted Apple's HIG as the basis of my interaction patterns, and innovated when our needs fell outside existing solutions. Most importantly, because of the first prototype I helped build, our company secured a round of funding, which let us double our team size.
What I learned:
This wasn't my first startup, but I learned a lot working with people who had such different priorities and backgrounds. I learned to adjust my concepts in order to meet engineering constraints. I learned how to tailor a presentation and quickly pivot to suit investor feedback. As a designer, doing rapid iteration in high fidelity turned out to be really time-consuming. In hindsight, getting buy-in from all the different partners would have exposed issues and flagged product issues much earlier in the process. That would have saved me hundreds of hours redesigning screens in high-fidelity, and given me time to validate ideas outside the studio. It was a grueling experience, but because of these mistakes, I learned to iterate in as low-fidelity as possible, getting everyone thinking about the problem sooner rather than later.
From a visual perspective, I think I would have benefitted from designing UI components and interaction patterns as a system, rather than in response to rapidly changing needs. Also, decisions like attaching colors to event types put an artificial cap on the number of types of events we could support before outgrowing the design pattern.
The simplest user flow walked users through a list of events, let them curate the type of events they wanted to see, and then see the details for events they selected. More structured product thinking could have helped us predict natural next steps earlier in the game.
Meet the team.
Abextra Technologies was founded in 2010 by M. Hooper and J. Williams. Allen Loggins provided creative direction, product design, branding, iconography, and production design. J. Lockett provided design support. D. Bretl and J. Nichols were front-end engineers. V. Menon and P. Katsev were backend engineers.
