So the past nine months of designing Glance have been a story of iteration…
Designing a piece of software to be intuitive and elegant is blood, sweat, tears, and a splash of carpal tunnel. There’s a hundred-and-one functions that you need, and it has to be well-organized, seem simple, be easy-to-use, and look good to boot. I’ve always felt it’s like cooking.
Out of all the projects I’ve designed for since starting as a UX designer (ooooh so long ago), Glance has proven the most difficult to design, mainly because there are no well-designed precedents for software in this area that are as complex as we’re shooting for. That, and the fact that the audience for Glance isn’t tech-savvy urbanites fooling around with a game. The audience is not just low on the tech-savvy scale, but they have genuine, understandable fear about using technology in their process. This is their research data, their lifeline, their everything. There’s no way they want to jeopardize that data in any way. So while there are many apps out there that give a nominal edge over an older technology, and that works for them to get new users, our app has to be so above and beyond, so friendly and unintimidating, so clear and understandable, that our audience will let go of the technologies they’ve been using for decades in order to take a chance on us.
Easy peezy. (not.)
It didn’t help that my iterations were consistently dozens and dozens of windows long, either. (I’m referring to all the little boxes you see in the pictures above.) I knew I was missing something, some clear path to guide the boat through the reefs. I had gone over many of the iterations with a large number of researchers, non-profits, and students, and they had all nodded along the way, but it wasn’t right. It wasn’t clear enough, I still had to explain too much, and there wasn’t enough surprised pleasure on their faces.
Over the last few days though, a number of events came together to provide for a break:
1– I took a long break of a few weeks from even looking at the UI, and instead spent a great deal of time thinking about why I was doing this and where I wanted to put my energy in the project.
2– Necessity propelled me to design an official logo for Glance (we just joined the local Social Enterprise Greenhouse‘s mentoring program, and so we needed a logo for their site) which forced me to confront the questions of: Everything I can spend paragraphs articulating about what Glance is, how do I boil it down to a visual essence?
3– I read Apple’s IOS 7 design guidelines. Say what you will about it, those designers put an incredible amount of thought into the new design and how the visual and motion design reflects the information design. They got rid of a lot of the old useless crap like skeumorphism, but finally took the next step past flat UI into layered UI, which I had been hoping metro would do but didn’t. (For a great article on that, check out this article on PencilScoop.)
Between high-level priority and identity consideration, the pressures of a deadline, and relief from knowing I had more flexibility in how I wanted to design than I had in IOS 6, I had a few breakthroughs.
First off was the first draft of the logo and aesthetics:
And then, with those aesthetics and that abstract understanding of what Glance was about fresh in my mind, I put all my Illustrator files away, went to my little notebook, and started doodling UI flows.
You can’t see anything in that picture, I know. What, think I’d ruin the surprise now, before the app is out? Psssh.
But my point is that all of it simmered down to two scribbled flows and about ten UI windows each about the size of my thumbprint. The navigation structure, which had before been long and complex, the functionality which before had been like marbles all shoved in a box, had condensed after months of stewing and stirring into something which gave me that wonderful feeling of being “on to something”.
I couldn’t have gotten here if I had over-designed my first iterations. At last, the careful approach I took months ago paid off, by leaving myself still willing to make large changes!
I’m sure it’ll continue to evolve, and I’ll still get things wrong, but there’s little to compare in satisfaction with making an elegant solution for a complex problem.