Process. Blegh. To some of us it’s a dirty word, especially where overburdened process has handicapped speed of execution and sapped your team of agility and excitement. I hear this sentiment time and again as reason #1 for big company designers looking to join small-shop startups to explore something new.
Yet process, when applied at the right time, in the right amount, can help you avoid hours of duplicate effort, improve team communication, and materially increase productivity. At Lyft, we’ve experienced a wide swath of the spectrum, from zero process in the beginning, to overcompensation along the way, to the present day where we now enjoy a happy medium. How much process is enough? In short it depends, but here are a few lessons I’ve learned on our quest for process nirvana:
Process, when applied at the right time, in the right amount, can help you avoid hours of duplicate effort, improve team communication, and materially increase productivity.
Process Should Scale
In the early days, the mobile team at Lyft was comprised of one designer, one product manager/designer, an iOS engineer, and a server engineer. Our development cycle looked something like this: (1) Hallway conversation with co-founders that leads to a napkin sketch/spec; (2) Capture tasks on stickies to track progress; (3) Generate UI mocks; (4) Implement; (5) Ship; (6) Repeat. A simple formula that worked for several months.
Because of physical proximity and small team size, staying in sync with teammates was easy, and we didn’t require much, if any, formal process. In fact, keeping process to a minimum made it possible to ship the first version of Lyft – from whiteboard to code completion – in three wild weeks.
However, as the company grew, it became increasingly difficult to keep track of priorities and progress while staying on top of new strategies, design requests, and dependencies. It was time to adjust. We tried on new processes and explored new tools, remaining flexible to pivoting when things were not working. More importantly, we purposefully kept our process super lean in order to keep moving quickly.
So as your team scales, consider a thoughtful application of process to stay organized and keep in mind that process shouldn’t be too rigid. Add just enough to smooth over the organizational hiccups without introducing unnecessary overhead.
More importantly, we purposefully kept our process super lean in order to keep moving quickly.
Ad Hoc vs. Scheduled Reviews
We like to move fast, and we insist on staying nimble as we hack our process and consider improvements thereof. So in the interest of speed, we’ve eliminated formal stakeholder reviews in favor of shorter, more frequent ad hoc drop-ins. Why? We found there was too much time spent scheduling the meeting, waiting for the meeting to happen, wrestling with AirPlay and A/V issues, and finally capturing and translating the input to do it all over again. Now, when any designer needs feedback or a green-light, we just grab the key decision maker(s), bring them to our desk, get the review done in real-time, and move forward. It takes a few cycles to establish a rhythm, but the quick drop-ins have been effective and best of all, they’re focused and efficient … and one less meeting, w00t!
We insist on staying nimble as we hack our process and consider improvements thereof.
A Narrow Focus
For more than two years, we were a design team of two that supported all UX/UI needs across the company, from mobile apps and emails to internal tools and occasional company swag. We multitasked like champs, but at scale, design was becoming a bottleneck. We staffed up and established a more reasonable designer-to-project ratio, but most of us were still spread across simultaneous projects in-flight. The context switching made it especially difficult to invest sufficient energy and focus on every project, and we felt we could be doing better work.
To solve this we decided to institute a one-project-per-designer policy, and do everything necessary to carry it across the finish within the given time constraints before kicking off another. Not only did this force us to focus, it also shined a bright light on process inefficiencies and ensured we were working on the right priorities. The end result has been higher quality output, less thrashing, a tighter backlog, and a happier design team.
We decided to institute a one-project-per-designer policy, and do everything necessary to carry it across the finish within the given time constraints.
Iterate and Evolve
While we’ve made positive strides, our process will continue to evolve. And being designers, we’re constantly iterating and tinkering as we’re naturally predisposed to do.
It’s worthwhile to embrace change and invest the time to tailor things to match your team’s specific needs. And rather than adopt process for process sake, or avoid it on principle, apply just enough and see where that leads. Make like a shark and keep moving!
by Frank Yoo
Frank is the Director of Product Design at Lyft where he's focused on building experiences that connect people and make transportation fun. His background is rooted in UX and product management, with prior roles leading design at LinkedIn and Yahoo!. He's a new dad, a husband, and principle frisbee thrower for the family Boston Terrier, Arial.