Six years after “Don’t Make Me Think”, and your Web site still sucks

If you ever wondered what brings a business to its knees in bankruptcy, then look no further than their Web site. Being in the industry, and in the know, I am no longer surprised at the severely heightened level of disorganization on the majority of Web sites today. This lack of organization online typically reflects the level of organization offline. However, that does not stop my almost seventy-year-old father from telling me he still cannot stand it. “Why do they have to make stuff so hard to find?” he often asks. Trust me dad, I can empathize.

Honestly, once you are privy to the bureaucracy, inner squabbles, and corporate misappropriation that happens during the development of a million dollar Web site, nothing surprises you. It is more often than not a miraculous occasion when the switch is thrown, because the creation that looked so good on paper, now looks like Dr. Frankenstein’s monster. It is a combination of body parts thrown together from almost every department in the client’s corporation, and the air is usually ripe with the stench of formaldehyde. Whispers abound, “This thing is too dead to revive.”

When Steve Krug’s book “Don’t Make Me Think” was published on October 13, 2000, it was like manna from heaven — a godsend, and real proof from a professional that things needed to change. The Man needed to hear what he wanted little to listen to, and across the land every CEO running the show had a free copy of this book placed anonymously in his mailbox. The Internet changed drastically in the years to come, and the User rose up from the ashes as a major player in the Web game.

Yet, without warning, and for some incomprehensible reason, the same abysmal mistakes have started creeping back onto major high-profile Web sites. We can now say with confidence to the new crop of CEOs, it is six years after “Don’t Make Me Think”, and your Web site still sucks. Not that I would dare to put myself on the same pedestal as Mr. Krug, but in the interest of saving face for a few big wigs still willing to change, I offer up the advice below. If you are in a position to demand change, or even to suggest change, then be not afraid. Nothing is more rewarding than suggesting your company’s Web site suck less.

The following are my top three suggestions:

Who cares what you want — the User wants what you sell

Too many companies are looking for that golden ticket. They want a winning idea that will drive traffic and sales up online. Yet, instead of finding out what the User really wants, they gamble with half-cocked ideas dreamed up by a brain trust. However, this trust, is often influenced by several other high-level decision makers, and the User is lost in the mix. The most important question that should be asked, is, are the users getting what it is that they want? To the brain trust, I say, no one cares what you want, because what you want invariably makes your Web site suck.

Selling on eBay does not qualify your marketing VP to run your Web site

I am convinced one of the primary reasons some of the most profitable companies offline cannot turn a profit online is because the individual calling the shots has no real Internet experience. That is, other than email and an eBay account. The exploration years are over, and long gone are the days when the Internet was fresh enough to justify learning on-the-job. Success requires experience, and that means verifiable success on the Web. Promoting your inexperienced marketing VP to be your Internet VP, and into a position of power over your Web site (especially if it is e-commerce driven), is a big mistake. It will make your Web site suck.

If you can only afford a single body, then you better make it an Information Architect

With respect to the program managers, programmers, interface developers and designers out there, if a company can only afford one body to make a Web site suck less, then it should be an information architect. A reorganization of the information appearing on a Web site can achieve major results with just a little fine tuning. An information architect or a usability expert can identify key aspects of a Web architecture that need improving, keeping the User in focus. An information architect can also design new components and wizards utilizing already leveraged technologies. Depending on the experience of the information architect, data can also be analyzed for patterns, mining crucial nuggets of information.

Conclusion

I suppose if you cannot even afford an information architect, then just do what Steve Krug suggests — buy his book.

  • Posted by Brian
  • October 22nd, 2006

Leave a Comment

Comments are moderated. No profanity. Only <a>...</a>, <blockquote>...</blockquote>, and <code>...</code> are allowed.

Seperate paragraphs by pressing the "Enter" key twice, or press it once to break to a new line.

18 Comments

#01, Oct 23 2006

Devin Ben-Hur

Well speaking of websites sucking and designers not having clues yet, what’s up with this?

layout.css:

body { ... font-size: 75%; ... }

Respecting Users includes presuming that they know how to set the font size they’d prefer to read text on a screen with. I like 10-12pt fonts, I don’t like 7pt fonts, so why would you choose to make everything on your page impossible for a normal human to read comfortably?

Don’t Make Me Fix Your Site!

#02, Oct 23 2006

Brian

Devin, the first thing to consider is that the design for my blog is one of the most popular blog design templates on the Web (not designed by me). The second thing to consider, is that the font size is percentage based, and therefore, relative. That means, at your discretion, you can change it within your browser — whether that be IE, Firefox or Safari. Changing the font to 10,11, or 12pt would assume that all the rest of the users of this site want that font-size. If you are a designer, you should take the time to understand CSS better before you make comments.

#03, Oct 23 2006

Alex

Brian, are you deliberately misunderstanding Devin? It’s pretty obvious to me that he means you should have left the font size at 100%, which in his case is 10-12pt. Why make all the text 25% smaller than your visitors preferred size?

#04, Oct 23 2006

Brian

Let me emphasize again that this blog design template is not my own — it was created by Sebastian Schmieg. Second, the fact that the font-size is 75% of Devin’s “preferred” or set font-size could never be a consideration in developing/designing a Web site. How do I know another user doesn’t prefer 14pt or 16pt or 11px?

I’m making an assumption, because I don’t know for sure, that Sebastian chose 75% because in the majority of browsers, 75% of the default browser font size looked good and is readable to the vast number of users. As I noted to Devin as well, the font size is relative, and he is obviously saavy enough to size it up in his browser settings.

#05, Oct 23 2006

Sean

If this was Sebastian’s reasoning, it’s pretty awful. The user set a preferred size because it was the one he was comfortable with. In many cases it may be the smallest size that he can read given his eyesight. In other cases — such as on the Mac — under a certain size antialiasing may be turned off. Setting the default font size to 75% is rather worse than hard-coding a front size: instead of saying you know best size-wise, you’re essentially telling _every_single_reader_ that, screw them, you’re going to make the font smaller than they wished, or perhaps needed, no matter what size 75% turns out to be!

#06, Oct 23 2006

Alex

> How do I know another user doesn’t prefer
> 14pt or 16pt or 11px?

You don’t. And you don’t have to know, because they can set that themselves. All you have to do is leave it alone.

#07, Oct 24 2006

Daniel

You really should look at this argument from a non-defensive position, your preaching about what is basically usability, but then argue against it on your own site.

>>he is obviously saavy enough to size it up in his browser settings
True, but I believe Devin’s, AND Alex’s points are that they, as users, shouldn’t HAVE to resize THEIR font.
Your site design should be text agnostic, at the very least the main body text should fit what the user wants, sizing up and down for individual and specific emphasis.

The person who designed the template, or that set the font size is completely irrelevant to the argument, because it doesn’t change the issue that the font IS smaller than the user wants.

Krug’s book (I own one, hand me down from a former employer) is more than about where information is displayed, it is about HOW information is displayed.

Just some food for thought.

#08, Oct 24 2006

stefan

I’ve set my font size to 11pt. But you think you know better and scale it down by 25%. Fix your blog first.

#09, Oct 24 2006

alles klar

before seeing the comments i had increased the font size.

May be in theory it’s good to have this 75% setting but in practice most end up changing it.

#10, Oct 24 2006

Brian

I appreciate all the feedback, and it is an interesting discussion, but let’s consider this in light of optimum usability and best practices.

Sean, hard-coding the font size as oppossed to a percentage based, would actually be a poor choice. You have then locked those users with IE into one size, and they no longer have the choice. Percentage based, being relative, is still resizeable cross-browser through the file menu bar. The fact that I don’t know an individuals setting, is the exact reason why you have to start with a baseline (default) value and work your way up or down.

As far as users “setting” their preferred font size — among the average user, you will rarely find those who know how to do so. Instead, starting from a blank slate — e.g. — the default settings, is always a better practice. If you take those default settings, and shrink them by 75%, then the text is still readable to the vast majority. If you have modified your font, and 75% is too small, then the percentage allows for an easy increase. In Firefox, it would be as simple as ctrl and the plus sign.

If you search several sites that use relative fonts, you will find that they do often begin with 100% on the body, but shrink down that percentage depending on the tag. So, whether that 75% appears in the “body” tag or the “p” tag matters little. It is common practice.

#11, Oct 24 2006

Dirk

If you start with font-size: 62.5%; then you can avoid difficult math when converting em to pt for the rest of your CSS. You can then specify everything in em (which is good practice anyway) and know at a glance what the default sizes will be.

1em = 10pt
1.2em = 12pt
1.6em = 16pt
etc.

#12, Oct 25 2006

graste

I don’t know, why people are arguing about the 75%. I increased the font size as well for a better reading experience, but I don’t think, that this is much of a problem. And even if you set 75% on the body that doesn’t necessarily mean that you don’t increase it afterwards in your css like Dirk already pointed out (you could set the font size of paragraphs to a higher value etc).

But back to topic: I think you made some valid points. An information architect is the best way to go. Many pages just suck on usability and information architecture. They look pretty. Many with too small font sizes, but who cares if you use a proper browser (like Opera or Firefox). The thing is, that many times you are not able to find the information that you’re searchin for in the first place.

#13, Oct 25 2006

Jim Bracher

The blog post is an arguement for user centered design, and the value of web-experienced personnel to manage business websites.

The comments are arguments about default font size. And no one observes that the default font size for both both Firefox and IE is considered overly large by most people, and that the implementation of font size control leaves much to be desired from a user-centered perspective.

With regard to topic #1: The web is no longer considered quite a new and shiny as it was, and we’ve wanted more web-savvy people in important positions for years. Clearly that battle continues. Does anyone have suggestions on what has worked and what hasn’t?

As for fonts, there is no good solution, only less bad ones. While those of us who like the tech have long since leared how to set our default font sizes, most the world as not. I still get complaints from various family members about the font size on some sites, regardless of how often I’ve told them about how they might be able to change it. Any font solution has to consider that most people will use the defaults of both their browser and the site. And most of the visitors will be using the IE current version. Because of all of that, percentile specification of fonts is the least bad solution. While it annoys some of us, we have the ability to easily change the font size. And if you will annoy someone, it might as well be someone who has the ability to do something about it.

In conclusion - the post was a good reminder that we are still struggling to make the web better, and user-centered design is still the best idea. Thanks.

#14, Oct 25 2006

Some real user

A lot of tech people have no clue about just how technically inept most net surfers actually are. Go ask around on e-bay or some other site that attracts a lot of non-techies and see if any of them have any clue about browser font sizes or re-sizing/setting them.

My mom would be like “what’s a default font size?”
Most of my co-workers would have clue they could resize fonts. All use MSIE defaults.

#15, Oct 25 2006

m3nt0r

OT: Why people judge someones thoughts upon their website they even haven’t designed by themselves is beyond me. /2cents

Good read, thanks.

#16, Oct 25 2006

Carey

Sorry to interrupt the discussion started by the purist who forgets about the 90% of web users who don’t know about changing font size. But I wanted to say thanks for the article. I read Steve Krug’s book when it came out but you’ve got me reaching for it again. It is too bad that the function of Information Architecture is not better understood and appreciated by the people who set web site development directions. Users have at least two visceral responses to the web interfaces they encounter: one visual and one functional. Too often web makeovers spend a lot of energy on visual coherence but only lightly address the functional considerations of information architecture, producing great-looking sites that don’t work well for people.

#17, Oct 25 2006

Brian

Thanks everyone for getting us back onto topic :)

@Carey –
I think you hit the nail on the head. Clients reject the notion that their architecture needs to change because the difference is not often as noticeable and drastic as a face lift, or a new widget. It also requires a previous data set of user statistics and traffic in order to give hard proof for ROI. This is something that companies still do a poor job at — usually you get a mess of log files that need sifting through, and that doesn’t really tell you much of anything useful. So it makes it harder to justify spending money on a usability expert.

@jim –
One reason it might be so hard to get “tech web saavy” individuals into higher positions is because they enjoy too much of what they do — getting the hands dirty and all. Moving up the chain of command sometimes means letting go of what you really enjoy most, and so employees that have the know-all to be a great CEO opt not to in order to avoid the bureaucracy.

#18, Mar 28 2008

Mark Turansky

I think you’re spot on.

“Don’t Make Me Think” changed how I think. It’s more than just websites, too, it changed how I think about code.

I can often be heard saying (probably to my coworkers dismay) “Don’t make me think about your code.”

I wrote about it here:
http://blog.markturansky.com/archives/70