Usability vs. Ease of Use

In a Web 2.0 world, we want everything to be customized. We want our web applications to know who we are, where we are, what we like, even what we're doing right now. There are a few techniques that designers and engineers can use to get this information without asking us - IP targeting, for instance - but they remain rather blunt instruments, and often leave us feeling violated, as if an evil internet spy was lurking just beyond our gaze.

The solution is to put the customization in the hands of users, allowing them complete control of what information they share, giving them oodles of preferences and settings, and making sure they can change everything at a moment's notice. Seems simple enough, right?

The problem is that most users won't ever find all those switches and dials, let alone know what to do with them or what mix of this setting and that toggle will result in Goldilocks-style contentment ("a this privacy setting is just right..."). There are a few us who enjoy this kind of masochistic settings torture, but we're not average (or stable). For everyone else, there's the dreaded user set-up guide, a ubiquitous necessity that's found its way into every website you've ever signed up for.

At Pinyadda, we're all about customization. It's our bread and butter. But we're also committed first and foremost to making our user experience completely painless. The place where these two goals conflict most intensely is in the set-up process, and it's where almost all of our product attention has been directed lately.

The most difficult part of designing any component is finding the ability to put yourself in other users' shoes, to imagine what it's like to be anyone but yourself. This is multiplied almost tenfold in the setup process, and the consequences of getting it wrong are similarly heightened. If you can't set it up, you can't use it. Imagine how much business Ikea would lose if they made you put the furniture together before you bought it. That's the kind of thing we're dealing with.

So how do we bridge this gap between usefulness and ease of use? Between customization and speed of adoption? It's a tough nut to crack, and something that probably takes a few failures to get right. A few basic principles we've learned (sometimes the hard way):

- Make as few assumptions about the user as possible. When you work on something every day, it can be hard to make yourself think like a brand new user. It can almost never hurt to make things twice as simple. Then make them twice as simple again.

- There's no substitute for user feedback. It's tempting to stay inside the bubble and iterate on your own designs, but the only real metric you have for usability is users. Seems so simple, but it's always harder than you think to put your designs out there for critique. Try to remember the adage to "have strong opinions, that are weakly held."

- Speed is good, but quality is better. Brevity is important in a setup guide, but quality of user experience shouldn't be sacrificed to achieve it. Would you rather get to mediocre experience fast, or get to a great experience a little slower?

- Structure makes people comfortable. Telling users where they are and giving them a clear understanding of what's next is important. Showing the steps up front and giving clear feedback when each one is completed goes a long way toward helping people feel comfortable and relaxed instead of anxious and annoyed.

There are probably a lot more, but we'll stop there. What do you think? Can you think of a great setup guide and why you liked it? Do you have any pet peeves when it comes to setting up a new service, or any tips for how to make it great? Leave your ideas in the comments.