article options:      send feedback        download PDF

MCNews: What is it about web site design that led you to write your book on usability?

Krug: Two things, I guess: First, there's the fact that it's so darned hard to get it all right--to come up with a site that's really easy to use. There are so many pressures at work--ludicrous deadlines, flocks of stakeholders with conflicting interests, changing technology, etc., etc.--plus the fact that you're designing for users who themselves have very different objectives, cognitive styles, information-gathering strategies, background knowledge, attention spans, etc., etc.--that it's very hard to build a site that works flawlessly--or even well. So the people who are trying to get it all right really have my sympathy, and I feel like they deserve as much help as they can get.

And second, there's the feeling that so much of what I do is basically common sense--something that almost anybody should be able to do, to some extent. Of course, people like me who've been doing it for years and years will [probably] always be a little better at it and know more about it (I like the rule of thumb that it takes ten years to become an expert at anything), but I still feel that with some guidance people who are motivated to learn about it can get pretty far on their own.

And they have to. After all, even for companies that can afford to hire a consultant like me, I can't be there every day with everyone on the team as they make the thousands of large and small decisions that affect usability. So even if they hire me, I feel like most of my job is educating people to make the right calls on their own.

And most sites can't afford an expert at all, so a book can come in handy. (Personally, I know there are times when I absolutely need a plumber, and I'll pay [almost] any price for one, but a lot of times a good book on plumbing will see me through.)

MCNews: Can you explain Krug's first law of usability and how it should be applied?

Krug: The "law" itself is pretty simple: Don't make me think. I've used it for years with my clients, and it really means exactly what it says: Don't do things that force people to think unnecessarily when they're using your site. I find that most people are quite willing and able to think when it's necessary, but making them do it when there's nothing in it for them (other than compensating for your failure to sort things out properly) tends to be annoying--and worse, confusing.

When you watch a lot of people use web sites (which is what usability experts do), you realize that even minor things that are left unclear or ambiguous often lead users astray and keep them from succeeding at whatever they're trying to do on the site.

The best way to apply this "law" is by testing whatever you create, which is what the last third of my book is about: low cost, do-it-yourself (unless you can afford to have someone else do it for you) usability testing. The only way to tell whether what you've built forces people to think or not is to ask a few people to try using it, and have them tell you what they're thinking while they do (the so-called "think aloud protocol"). It's usually obvious right away whether you've succeeded or not, and what parts you need to improve if you haven't.

I tell people to look for the questions marks forming over their test users' heads. If you see one, the part of the page they're looking at probably needs to be clearer. You just test, tweak the design, then test again until the question marks are gone (or at least as many as you have time to get rid of).

MCNews: Are there two or three principles of site navigation that should be included in every site?

Krug: There are probably twenty important ones, but here are the first few that come to mind:

MCNews: You've written that the tag line and welcome blurb on the home page are great ways to communicate your message. What advice do you have to improve those parts of a home page?

Krug: The tag line should be next to your logo (or right below it) at the top of the page, and everybody should probably have one, not just big companies. Choose a tag line that actually says something about what you are ("The world's best source for ice axes"), not a meaningless aphorism ("Taking you ever higher…"). And if at all possible, find a tagline that tells me what differentiates you from everyone else--why I should be here instead of at some other site.

A welcome blurb should be short (shorter than you think, about 30 words maximum) and contain real content ("3,000 products," "200 locations worldwide," "24 hour hotline," "Free shipping"), not "motherhoody" mission statements ("Providing leadership and outstanding service in the world of cold weather climbing technology"). Start with your "elevator pitch," then boil it down even more. People are not going to read very much on your home page; they just want the gist in a short sentence or two. (If your message is complex, make it into a short list of bullet points.)

Of course, you may not even need a welcome blurb if your message/business proposition is simple and you can convey it clearly just by showing some examples of what you do or what you're selling.

MCNews: How can non-technical people select web designers who make sensible usability a key part of design?

Krug: If I were hiring, I guess I'd look at the sites they've done (hopefully comparable in some way to yours) and see if I can understand what the sites are for and find my way around them easily. You can't hold them responsible for all the sites' sins, though, since designers don't always have the final word.

Then I'd ask them to look at your current site (if you have one) and one of your competitors' sites, and give you a few thoughts about what works and doesn't work in the designs. The key point here is not so much what they're saying as whether or not what they say makes sense to you. It'll give you good clues about whether you're going to be able to communicate well with them.

Also, if they sound like they're more interested in creating a portfolio piece for themselves than something that serves your objectives--and your users' objectives--I'd keep looking. (Of course, personally, I'd also ask them if they'd read my book….)

MCNews: When you review a site, what is the most common problem you find?

Krug: Too much--of everything. Too many options, too many offers, too many words, too many things competing for my attention visually. (Which is not to say that the barrage approach can't work sometimes. It's just that it's usually not the best approach since it often leaves people struggling to figure out what's really important.)

But if you asked me tomorrow, I'd probably give you a different answer, since there are a lot of very common problems. (So many problems, so little time….)

MCNews: Any web sites come to mind that you think do a really good job with usability?

Krug: I feel like I should come up with a new answer for this every time someone asks me. But the truth is, it's still the same old usual suspects: Amazon, eBay, Google. I like BestBuy.com a lot, too, but I think I'm biased by the fact that I like their in-store experience so much. (They always seem to hire nice, smart people who understand that their real job is to help you solve your problem.)

The common thread with these sites is that they figure out something that people want, work hard to figure out how to present it clearly and effectively, and keep reinventing themselves as they learn from their experience. Of course, even these sites have usability flaws, which just shows how hard it is.

I wish more new outstanding ones were coming along all the time, but they're pretty rare. Again, it's just so hard to get it all right. You need a good idea, painstaking implementation, and the will to resist all the natural forces that conspire to make things overcomplicated and hard to use.

MCNews: Any resources or sites on usability that you think are particularly useful?

Krug: Well, besides my book, these days I like Hot Text by Jonathan and Lisa Price (about writing for the Web, which usually gets very little attention even though it's crucial--after all, how many Web sites can afford a writer?), and Homepage Usability by Jakob Nielsen and Marie Tahir--especially the 30 pages of guidelines, which are worth the price of the book by themselves. Jakob's various reports (nngroup.com) are very good, albeit a little pricey for people who have to pay for them themselves, but his Alertbox columns (useit.com) are free and always a great source of insights.

There's also Keith Instone's link collection Usable Web (usableweb.com), John Rhodes' usability blog Web Word (webword.com), the information architecture ezine Boxes and Arrows (boxesandarrows.com), the National Cancer Institute's collection of guidelines (usability.gov/guidelines) and--probably the least known, but one of my favorites--Usability News from the Software Usability Research Lab at Wichita State University. That should be enough to keep anyone busy for a while.

MCNews: Thanks for your insights.

Find out more about Steve Krug, his book and his services.