My favorite apps are the ones that do one thing, and do it well. The user experience of these single-purpose apps can be imagined as a hallway that the user travels down. There’s one way in (launching the app), and one way out (exporting a PDF of your screenplay in Slugline, successfully importing your camera media in Offload).
BulletProof had the exact same checksummed import and backup features as Offload. But it also had many other features. That’s good, right? Who doesn’t love features? You could color correct your footage, and add tags and other metadata. You could transcode to a variety of formats.
Here’s what that looks like:
It’s easy to argue that features are purely good. One might take the stance that they don’t get in the way unless the user needs them, and then they are welcome. But look at that hallway. None of those new doors are “in your way.” Yet most people will find themselves curious about what’s behind them. And what’s behind each door varies greatly. Preferences is a door with a few buttons, but Color Correction is a door to a ton of control and power. Playback is barely a closet. Transcoding is actually another exit.
Is there stuff beyond these doors you should be using? Or is it safe to ignore them? If you open Color Correction first, you might think that every door has a big, powerful room behind it. If you open Playback first, you might think each door is just a pantry.
It’s hard for us developers to realize that people are terrified of using our apps wrong.
When you start adding doors, you add distractions, yes — but worse, you create a sense of unease in the user. Your app is now a Skinner Box; the doors offering unpredictable rewards. Every door is something new to learn. Every door is an opportunity to use the app wrong.
If the user has any other way to get their work done — even a terrible one, they will respond to this feeling by not using the app. This is what happened with BulletProof. People didn’t use it because we gave them too many chances to use it wrong.
Too many doors.
Photoshop is All Doors
Think of your friend who is really good at Photoshop. The one who uses it every day for work. The expert. Call them up and ask them how to do something simple, like mask one layer with another’s luminance. I guarantee you they’ll say something like: “Well, I know there are better ways, but this is how I do it...”
“I know there are better ways of doing it” really means “I haven’t opened all the doors, so I’m sure I’m using Photoshop wrong.”
Eliminating Doors Means Saying No
I listen to a lot of podcasts. I’ve tried a few third-party podcast apps for iOS, but always came back to the buggy and annoying Apple Podcasts app. When Marco Arment, creator of Instapaper and The Magazine, released Overcast, it took me less than a day to switch over to it for 100% of my podcast listening.
The other podcast apps seemed to compete with Apple's free offering by adding more features. Overcast won me as a paid user with a couple of well-chosen features, but mostly with its minimal doors.
Marco hosts his own podcast called ATP, and has appeared on a few others discussing the development of the app. He’s been candid about design decisions both in these interviews and on Twitter. Maybe my favorite of his tweets on the subject was this one (since deleted):
“I made Overcast because I had a different opinion on how, why, and where to say “No.” I’ll always need to be careful not to ruin that.”
When people make totally reasonable feature requests of my tools, I’m always carefully weighing the potential benefits of the feature with the inevitable tax of adding a new door. This is a particularly important thing to consider when your app is popular largely because it’s a simple and elegant alternative to a feature-rich incumbent.
How do Apps Wind up Cluttered with Doors?
One reasonable feature request at a time.
I know it’s hard to believe, but if I ever say “no” to your feature request, it’s often because I’m trying to protect the hallway experience.
The obvious counterpoint to this is that there's always room for improvement in any app:
With Magic Bullet Looks 3.0, for example, we added many new features. But we held back on some because we weren't sure if they were worth the tax of another door. We also removed some doors, by eliminating little-used features.
Power And Simplicity
A simple tool that does a simple thing isn’t impressive, and neither is a feature-rich tool that’s difficult to use. But show me something minimal that enables a ton of creative power, and I get excited. The challenge of making something powerful enough for you to do your most important work, but door-free enough for you to enjoy the experience, is something I look forward to every day.