Blog

The occasional musing about design, games, music, etc.

Designing Through Problems

Often in the creative design/development space we find ourselves stuck on a problem, repeating ourselves and coming up with the same half-solutions over and over. Knowing how to identify these times when we’re stuck comes pretty naturally, but having a plan for moving your design forward takes experience. Here’s a handful of thoughts I like to share with other designers to help get them out of the conceptual mud. This stuff works for me when I’m against a wall — hopefully it will do the same for you.

Beginning — When you just don’t know how to get started.

  1. The first idea (and my favorite) for helping get started is the concept of Beginner’s Mind. This idea, borrowed from Zen Buddhism, reminds us to keep a mental attitude that is open, eager, and with as few preconceptions as possible. Try to think about how you’d approach a problem if you never knew anything about it or that problem space. This concept is generally described as “In the beginner’s mind there are many possibilities, but in the expert’s mind there are few.” Try to let go of your previous assumptions, but not your experience and insights!
  2. Next up is the concept of Mise En Place. Originally from French culinary arts, Mise En Place can be translated as “everything in its place.” I apply this to design as keeping my tools ready, updated, and handy. Keep your workspace (and your computer workspace) clean and free of clutter. Then finally spend some time making good templates and default layouts for your work. The idea here is that when the space around you is “ready to work” — so are you.
  3. I often find myself looking at a problem and wanting to dive deep down on the first interesting interaction problem I find. Often, diving into the weeds like this can cause me to lose focus on the big picture, so I try to remember to begin by designing in broad strokes. I try to design one full pass of the experience I’m envisioning and save the extra details for later.
  4. Fourth up for getting started is a trick I call fingerprinting. The idea here is to find another similar design and catalog abstract things about it. Describe how it moves, how it feels to use. Does it feel valuable or disposable? How much complexity is exposed? This helps you remove the actual look of the design and helps you think about what you actually find valuable about that experience. Then you can try to apply those values to your own design (without just doing what they did).
  5. Anyone who has collaborated with me on a design has probably seen me press for markers then pencils. This is something that has served me well for years. I find that fine details are not often useful in the beginning of a design, especially when discussing it with teammates. So I try to hit a whiteboard and draw explanations out with thick lines and little actual detail. Try to find the flow of the experience first. Once you can’t actually draw the detail you need with the marker — move to pencil and paper. This lets you “refocus” on the screens and re-balance them with this new detail added.
  6. Another nice way to help get you started is to just add constraints.Constraints force creative thinking. If you find yourself paralyzed by too many possibilities, add some hard parameters to your work (even if they’re not real constraints!). This can help focus your thinking and get you moving in a direction. Some examples of constraints I apply to myself are: setting a deadline, blocking out time for specific types of actions (sketching, wireframes, prototypes, research). Other times I find I may need to change my venue for a fresh approach to a problem. Booking out a meeting room or working outside can help shake things up creatively.
  7. My last tip for getting started is to just get to work. New designs don’t just spring from the head of geniuses, they take hard work and putting in the hours. If you can find some flow on a problem, ride it as far as you can, even if that means turning people away. I often will cancel meetings or just work from remote if I find some flow. It’s rare and amazing — don’t waste it. The more years of experience you have the quicker you can find your flow on a design. My favorite quote on getting to work is by artist Chuck Close — “Inspiration is for amateurs — the rest of us just show up and get to work.”

Progressing — Ways to get un-stuck.

  1. When I hit a roadblock at getting a design to evolve I often find it useful to think about the overall repeat usage feelings. I typically call it the 1st time, 100th time, 1000th time lens. This really helps me try to put myself in both a new users state of mind versus someone more familiar with the system, or finally someone who may be almost tired of the experience. These help me try on different points of view to refine my design.
  2. Applying a different sense of scale to your work can help you find what’s working or what’s not in a stalled design. Try taking your screen and doubling the amount of real estate you have. Or maybe half it, as if it were on a tiny phone. Rework the experience for those other sizes and try out some pro’s and con’s of each. What works at these scales? What doesn’t?
  3. When I get midway through a design and find myself a bit lost with where to head next, I try one of my favorite cloning techniques. It’s actually a very simple exercise that can really unstick your thinking. First take your current design and duplicate it. Then remix it and reshuffle it until it feels sufficiently different. Do that a few times then take the ones that feel the weakest and toss them (actually, archive them for later). This helps either find a new approach or help you validate that your initial approach is strong enough.
  4. A quick and easy test you can use to check on a current design is to toss your solution into some bad conditions. It’s more than likely going to end up there eventually anyway, but these can sometimes give you a peek into where the experience may degrade eventually. Demo the design on a phone or a low res projector to see where usability and legibility suffer. Trying old browsers or old computers is a good way to test the water too. This of course is no replacement for actual QA testing, but it can help get the creative gears moving in a pinch.
  5. Any designer worth their salt has a bookmark folder full of design pattern libraries that they can rely on to find common design solutions to frequently found problems. If you don’t, I highly recommend getting yourself acquainted with a few. IBMDesign Principles FTW, and UI-Patterns are a few of my favorites. One additional thing to remember about design patterns is that you should not just go to them as a default. If your design needs the experience to feel new, fresh, or innovative — a familiar pattern might not help so much. Always feel free to challenge a pattern when appropriate!
  6. When I find that an interface I’m working on is too noisy, or the actions are competing with each other too much, i’ll take a swing at something I call copy versus art.What your problem would look like if it was solved entirely with text? What if you designed with with no text and only iconography? Testing these limits can help you give visual priority to competing actions.
  7. The first entry in the The Laws of Simplicity by John Maeda (one of my personal design heros) is this: “The simplest way to achieve simplicity is through thoughtful reduction”. When in doubt, simplify.

Finishing — Getting to Done.

  1. One of the most direct ways to get a design wrapped up and ready to ship is to trim off all the non-essential (but probably cool!) extra bits. Typically referred to as a minimum viable product (MVP), this practice of taking an ideal experience and rolling it back to just the core essentials can really help streamline a design and launch faster to help get more realistic usage data back quicker. Just be sure your team is really bought in to the idea of iteration after launch!
  2. If your experience feels brittle or overly compartmentalized, take the time to kill as many dead ends as possible. You should know what every list, display, and screen looks like in a “zero state,” when there is no data to show. Think about as many possible error stats as you can and make them more useful. I’d recommend you don’t design for every single possible problem, but try to make the player feel like they didn’t break it.
  3. Sometimes all you need to wrap up a design is some fresh eyes to validatethat it’s good! Show your design to as many people as possible via over-the-shoulder testing, player labs, or invites to try out a prototype. I’ve been surprised how much just some reality-checking can help finalize a design.
  4. There is a real skill in knowing when you’ve “designed enough” on a problem. I think of this as knowing when to take your hands off the pottery wheel, because sometimes you’ll get this impression that if it doesn’t feel right yet you can just keep going and you’ll get it right eventually. When nearing the end of a project, just remember that in general more design doesn’t equal good design.
  5. Inevitably, as you approach the end of a problem you’ll get feedback from all sides. Peers, collaborators, bosses, clients, everyone. I always try to remind other designers that “You are NOT your work!” Take any and all feedback to heart, but remember it is about the design itself, not about you or your skill. Don’t try to judge your worth as a designer based on feedback on specific projects.
  6. One particularly painful part of the design process is when you have to kill off your favorite designs. You shouldn’t be afraid of dropping a design and redoing it if it means you’ll get a better solution in the long run. Just be sure you keep every approach in a scrapbook or design archive with some way of digging them back up later. I’ve had designs I worked on 3+ years ago become relevant at some later point and pulling up my old designs saved a ton of time and effort.
  7. Then finally, sometimes it’s ready before it’s ready. Occasionally we get a problem where your first attempt is strong enough to build consensus, launch, and start getting real usage data. This may seem crazy, but sometimes the best way is to just launch it and learn as you go.

So there you have it. Twenty-some tricks I try to use or recommend to other designers when they get stuck. These have been really effective for me over my career and many of the designers I manage and mentor have also found them valuable. These tips have been collected over many years of designing and collaborating with other very gifted creative people. Big thanks to them, and I hope you found these useful!

aaron