Alertbox: 10 Years

by Jakob Nielsen on June 1, 2005

Summary: 300,000 words of usability essays have had an impact: online user interfaces are considerably easier to use now than they were in 1995. Many predictions and recommendations have come true, though the full Alertbox vision is far from realized.

It seems like I wrote "Alertbox Five Years Retrospective" just yesterday, but another five years have passed. Since writing my first column in June 1995, I've published 247 columns comprising almost 300,000 words. That's a lot of writing and a lot of content to give away on the Internet. Has it been worth the effort?

My most-read article, "Top Ten Mistakes of Web Design," has long passed two million readers. The average Alertbox gets 300,000 page views; the total number of page views for all columns is about 50 million. Even if I never wrote another article, the Alertbox would be a 100-million-page-view project, because the next ten years will double the readership of the archived pages.

Yes: it was worth it.

When I conducted my first user tests of websites and intranets in 1994, I was probably the only person in the world with this esoteric interest. Web people didn't care about usability, and usability people didn't care about the Web. After years of incessantly promoting user research findings for websites and intranets, the situation has changed: thousands of people now work on online usability. Nielsen Norman Group alone has trained 11,208 people and, given that many other places teach usability as well, the worldwide total is no doubt much larger.


I included cartoons in the Top Ten Mistakes of 2002. It was challenging (but fun) to come up with illustrations for often-abstract usability principles. Even with concrete, visual principles, coming up with a humorous cartoon was sometimes tough, as when trying to illustrate "small font sizes" in a readable manner.

In addition to mainstream user coverage, my column content expanded to include findings from studies of children, teenagers, senior citizens, users with disabilities, and low-literacy users. Each group requires special considerations; as the Internet matures, designers must recognize that most users are not just like them.

I reached beyond websites and intranets to cover email usability, including both email newsletters and confirmation emails. I also covered many non-online topics, from consumer electronics to Harry Potter .

Of course, I continued covering many themes from the first five years, including an emphasis on simplicity, user empowerment, and usability methodology.

Predicting "The Long Tail"

In 1997, I wrote the " Do Websites Have Increasing Returns?" column, discussing the relative value of big and small websites. I predicted that small sites would generate 75% of the Web's total value because they can be more targeted than big sites.

Admittedly, my predictions weren't always 100% accurate: I assumed we'd have 100 million websites by now, and we still have "only" 64 million. But, my main point has now become received wisdom: the "long tail" meme states that the aggregate sales of non-bestsellers sum to much more than the highly visible business done by the few top hits.

Most current discussions of the long tail underestimate the non-hits and assume that each point on the curve has the same value. But on the Web, being small means that you can better target your content and thus provide higher value per unit than more generic services.

In a follow-up Alertbox, "Diversity is Power for Specialized Sites" (2003), I returned to this second part of my argument and emphasized the treasure found in online niches.

I should be happy that half of my position has become widely accepted and the driving force for heavy VC investments today. But there's still too much commotion about aggregators, and too little interest in niche services in their own right.

How Much Progress in a Decade?

Taking a widescreen view of the past ten years, there is no doubt that the Web is making usability progress:

  • Blatant design stupidities like splash screens are pretty much history.
  • Usability metrics are trending up, whether we measure success rates or guideline compliance.
  • Many companies now have dedicated usability specialists on staff, and even more companies say that they want to improve the usability of their websites and intranets.
  • Internet managers have become more willing to follow design conventions rather than flout them and thus annoy their users.

A decade is not much time in the great scheme of things, yet there has been significant movement in the direction I've argued for.

And, yes, there have been disappointments. Success rates increase by about 4% per year and now stand at 66% across corporate websites. Given this, eight years remain before we'll get truly usable websites. In addition, website quality level is horrendous relative to the standards of six sigma or any other industrial-strength quality engineering process.

When I was wrong, it was often because I was too enthusiastic about a new technology's potential. When I was right, it was often because I was conservative. One of my bigger mistakes was to expect rapid progress in mobile services. WAP was a deserved flop, but I expected the next generation to be better and more widely used in its wake.

Finally, there's a reason the old "top ten mistakes" articles continue to be popular: despite clear documentation, website designers continue to make these mistakes at an astounding rate. Sad. But at least the mistakes are not as common anymore. We really are making progress. Even so, I'll have plenty of fodder for hundreds of columns to come.

Neglected Columns

Several important columns from the past five years are not getting the traffic they deserve:

Share this article: Twitter | LinkedIn | Google+ | Email