Jakob Nielsen, Ph.D., is a User Advocate and principal of the Nielsen Norman Group, which he co-founded with Dr. Donald A. Norman (former VP of research at Apple Computer). Until 1998, Dr. Nielsen was a Sun Microsystems Distinguished Engineer and led that company’s Web usability efforts starting with the original design of SunWeb in early 1994.

Nielsen’s Alertbox column about Web usability has been published on the Internet since 1995 and currently has about 200,000 readers.

Amongst other things, Jakob Nielsen has been called “the king of usability” (Internet Magazine), “the guru of Web page usability” (The New York Times), and an “eminent Web usability guru” (CNN).

Love him or loathe him, his contributions to the internet debate, particularly on the subject of Flash, have made him infamous and raised the important issue of usability amongst web designers.

PIXELSURGEON: So, Pixelsurgeon… just going on first impressions, how do you think it rates in terms of usability?

JAKOB: The very first impression was a splash page, which always says to me that this is a site that cares more about image than usability. The highly rendered bar of soap is a pretty image, though I admit to not understanding how it actually communicates the purpose of your site.

On the homepage, the right-hand column of links is virtually unreadable and needs some kind of explanation for why each of the links is recommended. The main news area on the homepage has too many small boxes. The category pages for features and interviews are much easier to scan and provide a better overview of the available stories. The site has a lot of small pop-up windows, which can often feel intrusive and overwhelming, but some of the small interactive animations were easy to control and quite enjoyable to play with.

Several windows contained numbered lists of images without any indication of their content. Usability would improve by either showing small excerpts of each image or by giving each image a meaningful name as opposed to a number. You cannot assume that users have time to download all the images, so you need to help them choose. [Note: Jakob was referring to the previous Pixelsurgeon design]

How did you get to become “the world’s leading expert on Web usability” (U.S. News & World Report), “the guru of Web page usability” (The New York Times), “the smartest person on the Web” (ZDNet), and the man who “knows more about what makes Web sites work than anyone else on the planet” (Chicago Tribune)? And do you agree with those titles? Are there any other “usability gurus” out there you would happily share these titles with?

I reached my current level after twenty years of doing usability, including nine years of doing web usability, where I have been very frank and outspoken about the things we find when testing websites or other user interfaces with real people. Also, many of my early predictions or insights from this research have later come to be fairly broadly accepted. Not everything, of course. I definitely make mistakes, but I have kept every one of my articles about web usability since 1995 up on my website for the world to see. Very few other analysts dare to make their old predictions and analyses available.

For example, even though I recently changed my position on Flash because of improvements in the latest release, I have retained the full text of my original harsh article from two years ago. Very few people have the intellectual integrity to stand by their earlier comments, even when later developments cause them to change their position.

My personal goal is to be the world’s number one usability expert. The difficulty of this goal is that you have to rely on other people to judge you. So I am certainly happy every time I get positive feedback.

Besides the satisfaction of meeting my personal goals, there is a second reason to aim to be the best. Recognition is a powerful tool in defending users’ rights and getting project managers to listen to usability recommendations.

There are definitely other good usability gurus in the world. Don Norman and Bruce Tognazzini are two incredibly experienced and brilliant gurus, and I am privileged to have them as my colleagues and have the everyday opportunity to learn from them. I have great respect for Stuart Card at Xerox PARC who is the world’s leading usability scientist. There are many other senior usability people who are very experienced and talented but who cannot really be considered gurus because they devote most of their talents to their own company’s projects. Gurus need to change the world, and you do this by impacting so many other people that they will accomplish much more than you could ever hope to do on your own.

Why do you do it?

I hate difficult technology. It annoys me on a personal level to see designs that violate usability principles that we have documented years ago. Every time this happens, which is several times per day, I get motivated to continue my work and evangelize usability even harder. Computers don’t have to be this difficult.

Isn’t usability just opinion ” what’s usable to one person is unusable to another ” or is it based on cold hard facts?

If something is unusable to even one person, then it ‘does’ have usability problems and is not the perfect design. The question is simply how ‘many’ users will encounter the usability problem and how much it will hurt them. These two assessments are partly a matter of judgement and experience unless you are willing to invest a large budget in running a measurement study with a huge number of users. It can all be made into cold hard facts, but in everyday projects, we do rely on judgement and experience in prioritising the usability issues because the elaborate tests are fairly expensive. Once you have seen something cause problems in twenty different cases, you can be fairly justified in predicting that it is likely to be an issue for design number 21 as well.

How can designers create attractive interesting sites, and still expect everyone to see them the way they intend them to be seen if different browsers, after all this time, and talk of standardisation, STILL render HTML pages differently?

You cannot. All you can do is to optimise for the most common cases and then make sure that the design is still reasonably good for all the other cases. Perfection under all circumstances is impossible.

How do you feel about the fact that “…UseIt.com rated only a “C” grade (75%) in an evaluation of 40 usability points identified by Nielsen himself…”? (Source: Anthony T. Dunn)

Given the budget spent on the project, it’s a pretty good score.

How can you apply a set of usability guidelines to everything on the Internet? With the huge diversity of sites out there, all with different purposes and different target audiences  and we know that people are not robots and do not all think the same  what are your usability guidelines based on?

Usability principles are rooted in the human experience and not in technology. Thus, the most fundamental guidelines hold true for all interaction design, whether websites or nuclear power control rooms. These principles have been established through forty years of human factors research and will not change unless you are a very strong believer in generic engineering.

More detailed guidelines may indeed change from time to time. For example, it is a fundamental guideline to provide response times of less than a second for users to feel freedom of movement in a navigational design. This might lead to a guideline to make web pages smaller than a certain number of kilobytes as long as most users connect through dial-up modems. This guideline would certainly change once broadband becomes more popular.

Then, there are guidelines that apply to a broad set of designs, but not to all designs. Examples include my 207 usability principles for e-commerce sites and my 32 guidelines for PR sections of corporate websites. These guidelines were derived from a large set of usability tests we have conducted with users visiting a range of designs that represent the specific type of site in question. Obviously, these guidelines would not apply to different types of designs, such as a set of movie reviews or a sales automation tool.

Finally, there are very narrow guidelines for vertical groups of websites. For example, there would be guidelines to increase the usability of flower sites that would be more specific than the general guidelines for all e-commerce sites.

I tend to only publish the first two types of guidelines: fundamental ones that apply to all user interfaces and general ones that apply to a broad group of designs. I agree that we also need more specific usability guidelines for targeted designs, but unfortunately it doesn’t make sense to publish these specific guidelines for vertical designs since there would be very few readers of such books or reports. Narrowly targeted guidelines should be written as internal projects by those companies that serve those markets.

What do you think of the idea that your usability guidelines suggest an almost “template” approach to web design, discouraging individuality?

Templates actually encourage individuality where it matters, which is in task support and content. The more companies can be assured that their design will work for users, the more they can focus on finding out what their users need and spend their resources on designing features and content to meet these needs.

I notice the text in the title tags of useit.com does not fit into a Mac Internet Explorer window’s title bar it just says “useit.com: Jakob Nielsen’s site (Usability an…”. Were you aware of this?

No, but at least all the main keywords are visible. I would have preferred if IE would have cut off the “an” rather than leave those two characters hanging.

It’s been a while since you wrote your infamous “Flash: 99% Bad” article. With comments such as “Flash content tends to be created once and then left alone” and “Flash content is typically superficial”, do you feel it’s outdated?

Unfortunately, there is still a lot of superficial Flash content on the Web, and I also think that it is true that many design projects tend to forget about Flash content once it has been created. Maintaining content and keeping it fresh are two of the main ways of supporting user needs and driving traffic. Still, it has helped that I wrote that article two years ago, because the negative trends are not nearly as strong as they used to be.

With all the criticism from Flash designers that followed the article, have you backtracked on it at all? Do you not think that perhaps it’s more a case of because Flash can be so easy to use, a large number of people who use it are using it badly? Blame the workman, not the tool?

This is true. There are certainly bad programs written in Java and bad articles written in English, and we would not necessarily blame the languages. Still, two years ago the Flash community did emphasise fluffy design and was not very interested in designing useful tools to support real user needs. I believe this has changed now, partly because I sounded the wake-up call that made many companies reluctant to degrade their user experience.

Are you warming to Flash at all with each new release, and especially Flash MX, with its plethora of new features?

Yes. Macromedia has taken the criticism to heart and has changed its strategy: the new Flash MX has several usability features that were lacking when I wrote my original article two years ago. Macromedia is now targeting Flash as a construction kit for Internet-based applications, which is a great way to get away from the tendency to superficial content.

In fact, Macromedia and Nielsen Norman Group have formed a partnership to develop the usability guidelines for Internet-based applications built in Flash MX. Once we publish these guidelines, Flash designers will be better positioned to deliver functionality over the Web in ways that empower users rather than confuse them, as can easily be the case when you start adding features.

You have been quoted as saying “the use of Flash typically lowers usability”. Although Macromedia will undoubtedly be biased when quoting specific Flash statistics that state certain businesses that have replaced their HTML sites with Flash sites have reduced bandwidth requirements, decreased page load times, and increased hits, when implemented properly, would you ever agree with the idea that Flash is better than HTML at making a site usable?

This depends a lot on the purpose of the design. If the main goal is to read articles, then I think that standard web browsing will often be best because it allows users to focus on the content. Some other day, I will complain about the lack of navigation support in current web browsers, but they do have the advantage of behaving in a consistent manner for all articles, and consistency is a great plus for usability.

On the other hand, if the design is very functionality oriented, as would be the case for a feature-rich Internet-based application, then I have long said that it is foolish to present it as a series of web pages. This is where Flash has the greatest potential for improving usability: when people need to use new features that are not provided by an article-reading tool like a web browser.

There is an intermediate case, where users are mainly reading articles but also need multimedia content. Here, Flash designs can do good, but it will be important to emphasize simplicity and provide the absolute minimum number of features so that the users can concentrate on the information and on getting on with their browsing.

A lot of sites on the internet are solely intended for quick and easy access to information, such as news sites, business-to-business sites and portals, but a large majority of others are up there primarily for entertainment purposes. Surely in these cases, navigation that encourages exploration and interactivity in a way that the user is not familiar with enhances the browsing experience? Can “different” not be equated to “interesting”?

No. Even when the content is entertaining, the navigation and other user interface elements need to be simple and get out of the way. We recently published a big report from a series of usability studies of children using kids’ sites (see “Kid’s Corner: Website Usability for Children”), and even though they liked funny sound effects and roll-over animations, the children were just as annoyed as adult users when it was difficult finding their way around the website. The kids wanted to get to their games; they didn’t want the navigation to be a challenge in its own right.

A large number of highly skilled, professional, well-respected and award-winning web designers are in opposition to some of your theories, especially those on Flash. Do you think there’s room for more than one set of opinions what defines a site as being “usable”?

Ultimately, of course, the users’ performance defines what’s usable. If users can easily accomplish what they want to do, then the design is usable. The main question becomes the extent to which one is willing to prioritise the needs of various classes of users, including, for example, those on low-end machines or people who are not highly committed to spending a lot of time on your company. Somebody who would fall into both categories could be a business executive dialling in on a laptop from a hotel room the night before an important meeting.

With your theories on web usability, are you ever talking directly to the businesses who want a website designed, as opposed to just the designers who are building the sites? Because more often than not, the client has a preconceived idea on what they think they want in a website, and no matter how good or persuasive the designer may be, sometimes the client just won’t budge.

I agree that many clients are clueless and ask for designs that would hurt their business. The beauty of usability is that we can show why this would be bad by getting test data of real customer behaviour. In my personal case, I am pretty privileged, because I am usually only asked to advise companies that have already decided to place high priority on usability.

You say in your book Homepage Usability “In contrast (to a sea of magazine covers at a news stand), users don’t see the homepage until they have already decided to single out the website and visit it. Thus, homepages don’t need to stand out and grab the user’s attention because the user will already be looking at them”. What about when a company’s homepage is brought up in a search engine alongside all their competitors’ homepages? Surely the user will open up a couple of the results, have a quick glance at each of them, and then based on their first visual impressions, shut down those that would appear to be less interesting? Seemingly in contradiction to the above statement, in the very next paragraph you go on to say “when the first impression isn’t good, you don’t get a second chance because the user will never return”, and “they first give the piece a quick once-over to see whether it interests them at all”… Perhaps the ‘splash screen’ is a good thing after all?

No, splash screens must die. They give the first impression that the site cares more about its own image than about solving users’ problems. It is true that a site needs a homepage that immediately communicates what the site is about and what users will get out of visiting, but one of the most important feelings to communicate is that of respect for the users’ time. People need to feel that a site is there for them and that it will be easy and worthwhile to spend more time on the site, or they will leave. An attractive visual design helps, as does one that prioritises the available information and features and guides the user’s eye toward the most important stuff. Still, this is a different goal than a magazine cover design, which only has to communicate excitement.

People know that they will be able to operate the magazine’s user interface, so a magazine doesn’t need to communicate ease of use on its cover. In contrast, one of the biggest contributions of a website to a company’s brand reputation is the ability to increase the score for “it’s easy to do business with Company X” on the annual customer satisfaction survey.

“I don’t hate Flash per se,” (Jakob Nielsen) claims. “It’s just being used in the wrong way and often gets in the way of the user’s experience”… (Source: Revolution Magazine). Surely the same can be said for an equal percentage of standard HTML sites?

Yes. I have also been very hard on many HTML sites. Not to worry.

You say on your site useit.com, when explaining why your site has almost no graphics, “Users do not keep their attention on the page if downloading exceeds 10 seconds, corresponding to 30 KB at modem speed. Keeping below these size limits rules out most graphics”, which may be a very valid point. But another very valid point is that users do not keep their attention on the page for very long if it looks very boring, and useit.com, with its reams and reams of uninterrupted text isn’t the most interesting site on the internet to look at. I’m sure your visitor stats show a phenomenally high number of visitors, but your site’s reputation precedes itself. Surely you can’t expect other lesser-known companies to follow this idea?

True, this particular site looks boring, but that’s actually a benefit since its only feature is one new article every two weeks. Looking severe is in character with the main message of the site, which is to get out of the way of the users and get them to the information they want. At the same time, I admit that many graphic designers might decide to skip a site that looks ugly under the assumption that it is not relevant for them. This would be a shame since I believe that good graphic designers can make substantial contributions to usability if they understand the main principles of the field.

Interview Links

You may also like