My Design Process - Part 3


You’ve discussed, debated, designed, prototyped and user-tested. Time to build! Yeah, you too. I’ve already written about the real reason designers should code, but here’s another: you’re about to wish you could. As your engineering pals start putting this product together, they’re going to uncover an untold number of corner cases you never accounted for in your design. They’re going to come to you and ask, “Hey, what happens when you have a digital order that’s customized, but also purchased from a Wedding Registry?” You’re going to stare blankly for a second, half your brain processing the request, the other half waiting to see if they’re kidding.

They’re not. Your layout is broken. And while you could sit there and verbally explain every little tweak and fix, life is much simpler and, frankly, happier if you just jump into the code yourself. Some of my greatest days have consisted of just passing front-end code back and forth with my engineering partner. It makes you appreciate the intricacies and borderline-insanity of developing complex systems. You’ll run into some nested if-statements and start questioning your own existence while unravelling just what the hell is going on. It’s great.


Once you’re code complete and have tested your work (hello, IE8, my old friend), you and your team get to press that big ol’ deploy button (not allowed to deploy to production? We’re hiring at Etsy). Hopefully you’ve already had your entire company dog-fooding your new release during development. If not, it’s time. Get everyone to use your product under real, live conditions. Gather feedback. Make adjustments. This also serves as a great way to triple-check that everyone who needs-to-know knows what you’re about to ship. Is support prepped to answer questions about your new thing? Is your community team ready to engage your most passionate users with confidence?

Sure, you could put this solely on the PM, but a big part of a designer’s job is communication, and that doesn’t just apply to your users. Get involved. Go with your PM to help explain UX details and field questions about the choices you made. Get your company excited about what you’ve been doing.

Write code, deploy to production, cheerlead your project. If it sounds like I’m telling you to get involved in everything, I am. The best designers I’ve known can’t stand being left out of this stuff. It’s no wonder their work is so amazing: there’s never a point in the process they’re apart from it.

Watch the Graphs

At Etsy, we have pretty sophisticated (in my experience) data analysis tools. That doesn’t mean you have to work at an established company to start working with data, though. There are great tools out there from MixPanel to Kissmetrics to, sure, even Google Analytics. At Zoosk, we ran off Google Analytics for quite some time (heck, as far as I know, they’re still on GA). When you’re small, just pick a tool and go for it.

The important thing is to have a measurable hypothesis (you’ve preferably gotten that in the early design stages so you have something to design against). Is the primary goal to increase the rate of sign-ups from organic traffic to your homepage? Let your experiment run for a little while and check in. Is it doing what you expected? Did it wash out? Did you actually hurt the number with your changes? Regardless of the result, ask yourself why it might have happened. Form a new hypothesis. Tweak your design and relaunch.

The most important part here at the holy-crap-this-is-live-right-now stage is to not let yourself get discouraged when you don’t get the win you wanted on your first try. Accept that, more often than not, you’re going to be wrong and that you’ll need to iterate your way to success. You could spend a year designing/redesigning a feature and you’ll still get surprised by real-world data. Don’t spend a year. Get to your first failure quickly and move on from there.

Wrapping Up

The truth is that these aren’t all the steps you could possibly take when designing a product. Conversely, it’s certainly neither necessary nor appropriate to go through each of these stages for every single project of every single size and scope. By trying out new processes, you can start to form your own, personal design process and find your own preferences for various situations. You’ll also, in a meta way, iterate on your own process your entire career as new tools become available and our old tools evolve.

The most important thing is to keep your focus on the user. It’s all too easy to get bogged down in the process and lose sight of why we invent these processes in the first place. What does it take to provide an amazing experience that helps your users accomplish their goals and enables your product and company to flourish?

Do that.

Don’t like visiting web sites? Sign up for the newsletter to get blog content and links I’ve found each week.


Now read this

Curiosity Required

Once, while looking for a new gig, I met up with a couple of potential coworkers for food and drinks. We were chatting about their project, what I’d worked on previously, etc., and eventually started talking about recent tech news. At... Continue →