Last week Valleywag posted a critique of usability guru Jakob Nielsen, who is apparently now taking some flack for his ongoing criticism of “web 2.0″ design. “Today, though, Nielsen’s dismissal of Ajaxy user generated content comes across as more crank than critic,” the article notes, taking issue with his “garish yellow-and-white design” and “tiresome refusal to use images to convey information and ideas.”
This editorial interested me because I struggle with the same issues as a designer at Google. I’ve watched the web 2.0+ world evolve around us, but our design mentality remains rooted in the early days of the web. Black text, white background, blue hyperlinks. And it’s not just Google and Jakob Nielsen – this aesthetic persists all over the web, despite many designers’ attempts to banish it forever.
Speaking for myself, the sites I use every day are still very much in line with Nielsen’s view of good usability. Google web search, Gmail, Reddit, Wikipedia, Twitter. Black and white and blue all over. Why does it work?
The interface needs to stay out of my way
Every application has a function. Assuming that function is well-implemented and something that I need to use, I will want to use that application. The interface should only be a tool that helps me get to that functionality. The moment I have to think about the interface – even to marvel at how cool the drag animation is – the application has failed in it’s primary purpose. It exists for the function, not the design.
“Now Anne,” I can hear you saying. “What if the point of my application is the experience? My Flickr/YouPorn/Facebook mashup is super-slick and fun to play around with. I get like ten thousand hits a day!”
Okay, yes. These apps do exist, and should exist. They can be anything from neat toys to richly detailed digital adventures. I’m not really talking about these kind of apps, because a user isn’t really trying to do anything with them. They are there simply to experience it.
When you have problems is when you get the bright idea to make your practical application, say a webmail client or a news aggregator, and make it “experience oriented.” A rich, novel interface can certainly give your site impact and memorability. I might tweet about how cool it is and it will probably generate a lot of interest and hype. These are good things, but they reside in the sphere of marketing, not usability. Designing for the launch is a mistake. Getting people to log in on day one is only half the work; getting them to stay is the other half. You want them to come back every day and the novelty of the rich user experience is only going to last so long. When it wears off, can they do what they need to do on the site quickly and easily?
Yes, latency is still an issue
…unless your target demographic really is rich, technologically savvy people with fast internet connections in the United States. Hey, it might work if your margins are really high, but if you’re writing a web app, chances are good they aren’t.
Reddit is… well, Reddit, but I keep going back to it as a source of things to read because it does what it does and stays out of my way the rest of the time. It is literally just a list of links on a page, which is exactly what I’m looking for. I want a service like that to load instantaneously, because I’m likely to hit it many times throughout the day. Waiting for an image to load annoys me. Waiting for a page full of them makes me want to go somewhere else.
If your app is going to have a rich artistic element, it had better damn well be customizable
Imagine if Apple set your desktop image for you. You use your computer every day, and even if it was the most beautiful photograph in the world, you would get pretty damn tired of that image. This situation sounds ridiculous when talking about client products because it’s well known in that world that customization is an important part of a happy user experience. I don’t understand why web developers haven’t picked up on this yet.
iGoogle is a good example of how to do this right. This is an application you’re meant to keep open in a tab all day while it pulls in news feeds and tracks your email and whatever else you configure it to do. When they realized iGoogle wasn’t exactly the prettiest thing to look at, they added user-customizable themes with colorful, rich banner images behind the Google search box. Even better, the image changes throughout the day, so every time you click over to that tab to check your email you might see something different. The end result is that it takes a while to get sick of a theme, and when you do you can always just switch to another one.
User experience is not all about the clicks, but revenue is
You don’t have to take my word for any of this, you know. You can use any number of free tracking tools (Google Analytics and Website Optimizer are my favorites) to learn more about the behavior of users on your site and run experiments with the design. Usage statistics certainly can’t tell you everything – they can’t tell you if your users are happy, for example. And there are some pretty ugly and/or misleading things you could do that would almost certainly increase the number of conversions on your site. I’m not trying to tell you to do these things. The negative impact to your overall user satisfaction and brand would not be worth the increase in clickthrough.
Even still, you have to be running these numbers and aware of them if you want to be an effective designer. “But the green links make users feel good!” is not an effective argument if blue links increase clickthrough by 5%. ROI is not a dirty word; embrace it, and if you keep a record of the data you collect and the tweaks you make, you’ll have a lot of ammunition handy when talks of a redesign come up. The first question you should ask is “What problem are we trying to solve?” and the second is “How can we prove that we solved it?”
The moral of the story
“So Anne,” you ask. “Can I use images or what?” The answer is, it depends. Pretty graphics and colors and AJAX and slickness are all great, when they don’t get in the way of what your app is supposed to do. No matter how you approach it, things that are clickable should be easily identifiable as clickable. Things that are meant to be read should be clear and easy to read. I’m stating the obvious here because it is the precise reason the text + hyperlinks paradigm is powerful and shouldn’t be overlooked.
Just look at Wikipedia. In many ways, it is the perfect web application. How empowered do you feel when you use that site? You have easy access to almost anything you could ever want to know, and anybody who has used a computer in the past twenty years can figure out how to navigate it. Even my grandmother knows blue underlined text is a hyperlink. It’s a big flag that says “you can do something here!”
Now, I know you need AJAX and pretty gradients and buttons to make your webmail client usable and attractive. Do what you have to do to make it work, but at that point you might want to ask yourself if the web is the appropriate platform for that application.
Remember, design is equal parts marketing and usability; approach it as such. Be pragmatic. Don’t mindlessly decorate. And don’t be afraid to use what you know works – even if you are in danger of Valleywag calling out your “so 1990s” design. Own it!