Eric Enge Interviews Usability Guru, Jakob Nielsen
Published: September 24, 2007
Jakob Nielsen is a widely recognized usability guru. Here is a sampling of some comments about him in the press:
- "the king of usability" (Internet Magazine)
- "the guru of Web page usability" (The New York Times)
- "the world's leading expert on Web usability" (U.S. News & World Report)
- "one of the top ten minds in small business" (FORTUNE Small Business)
- "one of the world's foremost experts in Web usability" (Business Week)
- "the Web's usability czar" (WebReference.com)
- "the reigning guru of Web usability" (FORTUNE)
- "eminent Web usability guru" (CNN)
- "perhaps the best-known design and usability guru on the Internet" (Financial Times)
- "new-media pioneer" (Newsweek)
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 he was a Sun Microsystems Distinguished Engineer.
Dr. Nielsen founded the "discount usability engineering" movement for fast and cheap improvements of user interfaces and has invented several usability methods, including heuristic evaluation. He holds 79 United States patents, mainly on ways of making the Internet easier to use.
Eric Enge: The first thing I would like to talk about is some good common rules that people should use for navigation on their web site. For example, you recently wrote about, there being no OK button on an option box in an application you encountered.
Jakob Nielsen: Yes, that article was about a particular graphics application where there were certain settings you change, and the dialog box for making those changes did not have an OK button on it, so there was no way of getting out of it. The point here is that people are just used to dialog boxes that have an OK button, and probably also a Cancel button, as the ways you dismiss one of these dialog boxes when you are done with it.
That's the expectation, because it has been part of the user interface standard for graphical user interfaces since 1984 when the Mac came out, and then Windows adapted it as well, and many other interfaces too actually. So, it's just the standard that people are used to, and that leads to the more general rule of doing what people are used to.
Because, you don't want your customer to puzzle over how to operate your user interface, you just want them to focus on your product, on your argument, on your ideas, whatever you are trying to promote, and not on the mechanics, because the mechanics are hard enough already. And, if you then get in their way by doing things in unusual manner, then for sure you are going to lose a large percentage of people here. I think a common example on the web of this problem is designers who make up their own scrollbars in Flash, because they don't like the standard simple plain old scrollbars that we've had since 1984, and that everybody is used to.
Eric Enge: Right.
Jakob Nielsen: But, the problem is when you do make up your own scrollbars, and we've tested many, many, many Flash designs, and they have other issues as well, people just overlook them. We've also looked at sites, for example e-commerce sites that had product listings that you could scroll through and what many users would do, they would think that company sold the visible products and there were no others, because they didn't see the scrollbar. And so, this company in reality was presenting maybe 10% of their product line to a large percentage of their customers, just because they had to make up their own scrollbar. And so, that's really the main guideline, I think of all the many user interface guidelines for this, do it the way people expect.
Eric Enge: Right, a corollary to this would be don't fall in love with how smart you are.
Jakob Nielsen: Oh yes, exactly. And, that is in fact the real problem, because the people who are on a design team, they sit and ponder that design 8 hours to 10 hours a day, for months in a row, and so they get much more engrossed in it. In contrast, the average customer will maybe spend half a minute on a homepage, and then they are going to stay or leave. And so, the amount of time that a new prospect has to understand your website or understand any particular given page is nothing compared to the amount of time that you put into it. So, you may think oh, it's boring, the customer doesn't think that, the customer thinks oh, it's easy, and he just moves on.
They click on that what they want, move into the site, customize the parameters they want, click to the shopping cart, proceed to checkout, all those boring things.
Eric Enge: Yes, exactly. It's a situation where we actually do want to bore them, relatively speaking.
Jakob Nielsen: If people get very excited about the product, instead of your interface, that's what you want.
Eric Enge: Right. The other thing that the web designer and the website owner can't ever fully appreciate is the perspective of their customer and the various audiences that they have on their website, and, getting into that mindset seems incredibly important.
Jakob Nielsen: Yes, except it's basically impossible; the main rule here is you are not the audience, you are not the customer. I mean there is a reason that you are the one working on the website, and you are not the average representative outside customer. There's a very small number of exceptions to this where people are in fact designing for a very small niche of people just like themselves, but that's the exception. The normal rule is, you are designing for a much broader audience of people who are very different than yourself, who don't know as much about computers, who definitely do not know as much about that company, that website or that product line, and they are busy, and they are also looking at the competition, and so give it to them straight.
Eric Enge: Right. They are task oriented and want to get it done and move on.
Jakob Nielsen: That's exactly right. That's another difference is that, if you work on a project you start caring about it, and you think it's exciting and interesting. For customers, it's just one of ten things that they just brought up in their search listings, and they click on a few of them and check them out. The prospective customer is just not bought in yet, and it's your job to motivate them and tell them about the product. But, they are not willing to jump through hoops just to stay on your website.
The main method we have of overcoming this gap in thinking between the inside thinking and the external thinking is user testing. These are many ways to do this, but the No. 1 basic method is user testing, and nobody should ever put out a website without doing user testing. I mean it costs very little, and it's going to double your sales or double your conversion rate, and so just test it with a handful of customers. You don't have to do an extensive big study, just test a handful of customers, and you will see how they think about it and what they want, and what they think is difficult.
Eric Enge: Right. And, it seems like another one of the things that's important is that the navigation be very consistent, both at the global level, and also your local navigation.
Jakob Nielsen: Exactly. We actually often see that navigation jumps as you move through a website, that's a classic example of kind of less professionally done website that was built out in pieces. For a big company what often happens is that they hire different design agencies to do their different product lines and each one makes up their own way of doing it. Then, as you move around that site, the navigation jumps and that's a classic problem. You want to have it be consistent, and the same everywhere.
The global navigation tells you about the main areas of the website, main kind of utilities like customer service, or about us, or contact us, things like that that you always want to have. Then, you have the local navigation which is actually more important in most cases, because people tend to arrive at a neighborhood of information and want to move around in that neighborhood. It's not so often they want to move to something completely different on the website. So, most often people are interested in a certain type of product, or certain type of service, and they want to know about what you are offering that's similar, and what are the differences between product 2 or 3 similar product offerings which is often very poorly explained.
The ability to say I want something that's a little bit more powerful, I want something that's a little cheaper, I want something that will work for whatever my particular circumstances are, and then do local navigation to the right product is really important. Particularly as people arrive from a search, because the search engines are not perfect, and so they will often kind of project the user deep inside the website at some place that's not always quite right. It's probably about right, because there's going to be a page that's about the query that the user types, but it's not going to be exactly what they want often. And so, having local navigation, to find similar and related topics is really important.
Eric Enge: Right. What about the "People who bought this also bought" notion that Amazon made famous?
Jakob Nielsen: I think this is a brilliant way of doing it, and they have definitely sold a lot of extra books, because they did that. Often you can do it yourself manually, because often you will have an understanding of the product line, but something like Amazon, where they have millions and millions of products, they have to have the computer to do it for them; they cannot really manually build up all these links. So, they do it by computer, but small companies with a smaller product selection should probably do it manually and just decide on, well, if you are interested in this product, here are 5 or 6 other ones that you might want to look at as well.
That's a great way of enhancing sales, but also keeping or saving a sale, because the person may be looking at a product that's not quite right. If you point them to something that is right, then you will keep them, and otherwise they may never find it on their own and you would have lost that sale.
Eric Enge: It really keeps you within the theme of the thing that they were looking for.
Jakob Nielsen: That's right.
Eric Enge: Breadcrumb bars are something that I am fond of. What are your thoughts on them?
Jakob Nielsen: I think it's actually a great navigation aid. It's not the primary navigation, but it is a great secondary navigation, and it serves 2 main purposes. One is to give people a sense of place which is really important on bigger websites, particularly if you think about users who arrive from a search engine, or who have done a search on the site itself. It's almost like there was a transporter like in Star Trek, where they are teleported into the middle of this jungle: telling them where they are is very, very helpful.
That's one great thing it does, and the second thing it does is, because it's actually clickable the different links show the different higher level categories that you're in, and so that allows people to find other related products; and again if they were not at exactly the one they want, they can go and find other similar things.
Eric Enge: Right. And, if they're just basically clicking around and not transporting using the search box, they know where they've been.
Jakob Nielsen: Exactly, and you can see where you have been. The one thing I do want to point out is never rely on breadcrumbs as being the only approach, because it's nice to also provide that more global navigation that talks about things like the customer service and various other things that are not really part of the hierarchy of people's thinking, but at some point of time oh, I need customer service, and then that link should actually be there.
Eric Enge: So, let's switch a little bit and start talking about some on page elements, one of the things that you wrote about recently was descriptive taglines for the company on the homepage.
Jakob Nielsen: It's amazing how often taglines are not descriptive. I mean most companies do have some kind of slogan, but it's rare that they describe what the company does, and there's often some kind of very wishy-washy feel good statement that could have been put on any of a thousand companies' homepages, and so it doesn't really do any good. The main purpose for the tagline in terms of the web is to give people this one line conformation of where they are, what they can do, and what the site is about. That can then also serve as something that you can put into the page title, or the bookmarks, and which also will serve as the main link on a search engine listing as well.
So, if you can find a very compact way of describing what people will get on this website that provides a lot of benefits. On the other hand, a very general feel good statement is something people are just going to ignore, it's not in fact going to make them feel good; you are just going to make them feel oh, here is extra blah, blah, blah word count.
Eric Enge: What I have seen people do too is that they have some feature on their site and they pick some names for it like they plan to trademark this name and make it the next trademark like Coke. But, for most websites that's never going to work, because no one has the budget of Coca-Cola.
Jakob Nielsen: Exactly it is very, very rare. In fact, Coke and Coca-Cola are two famous names for the same thing, and that's just very rare. The average companies would be happy if people can remember one name, and that should be enough branding right there in terms of recognition, and then the rest of the branding should really be in terms of experience branding which is, what am I getting, what are you about, what are you doing for me?
That is better achieved by using plain language to describe the features, because if you make up your own words for them, that's in fact not going to be special branding that people are going to remember; because there's too many of those on the web, and you are not important enough. These are hard lessons for most companies to really appreciate, that they play a relatively smaller role, and people are only going to be on their website for a few minutes. But, that's how it works, and so you should provide something that people can get benefit from within a few minutes, and then they are going to come back, and then they are going to keep using you.
Eric Enge: Right. This applies not just to taglines, but the business name itself, and the names of your products, and names section of your site.
Jakob Nielsen: Yes. The more it's described in plain language, more people will know what they are going to get, if they go there, if they click there, if they use that feature. On the other hand if it is described in made-up words, new words, the less people will know what it means. One of the big conclusions out of so many years of studying how users engage with websites is that they just don't have time to go into all the possible links that are on a homepage or are on a given section of a website.
The web is very rich, and it offers many, many choices which is good on the one hand, but in the other hand also overwhelming, and it means you just cannot have time to do everything. So, the web user has to be selective, it's just inherent in the nature of the web. You have to brutally cut down on the number of things you do, because you cannot try all the choices, because then you would never get done. People have to cut out things. They tend to cut out things they don't understand.
Eric Enge: Many years ago people used to think in terms of 8 second rule to load a page or users would abandon a page. Now I have seen data that suggests that many people form their initial impression in less than one second.
Jakob Nielsen: People tend to spend about 30 seconds total on a homepage. So, it's a very initial impression, but then you can look around a little bit more, but still remember that 30 seconds is not a whole lot, that's maybe a time enough to read about a hundred words. If you think about the number of words that are on a web page, that's a tiny fraction of the total word count that the user is actually reading during the initial visit. So, when we do eye tracking, which means that we measure where the user is looking on the page, you see the eye kind of flicker around the page in scanning and kind of looking at some of the many different sections to try to gather what the page is about.
Then, they focus their attention on just one small area, where they think it's most likely to be useful for them, and then they will start reading a bit there. But, there's not a whole lot that people can get even in those few seconds, so you have to really be plain spoken and short, to the point.
Eric Enge: Presumably all of this also applies to the anchor text you use in your links.
Jakob Nielsen: Yes, it applies dramatically to the anchor text, because again going back to the eye tracking studies, you see the eye move over the beginning few words, and then very often skip down to the next one. And so in the Eye Tracking studies, the first 2 words are going to be red, much more than word #3 and #4, which again going to be much more red than any subsequent words towards the end.
So, the first 2 words are really where you have to get users, and that's why you shouldn't use generic feel good vocabularies like welcome to try our, or something like that. You have to answer what's in it for me, why should I click here; try to put some of those action oriented words to tell people what it's going to do for them. Try to put them up in the first 2 words and also in the 3rd word and the 4th word as well.
Eric Enge: Right. Given that the page title is what shows up in the search engine, wouldn't that same reasoning apply to the title itself?
Jakob Nielsen: It completely applies to the page title, because the page title really becomes a link when it shows in the search engine. I feel it becomes a link and a headline, because it's the thing that people read the most when they look at a search engine listing. So, people will mainly read that page title, that blue link that's in bigger type, and they will mainly read the first few words of it. Then, if that seems interesting, they will kind of relatively quickly scan through that little summary that's under the title. Then, they may, or may not, look at things like the URL for confirmation of where is this thing going, but the title is definitely where people look the most, and the beginning of the title even more so.
Eric Enge: Right. You may want your company name to be part of that title as well, alright?
Jakob Nielsen: I think it depends also on how famous the company is, and how much to emphasize it? It does go to this point about getting people a safe feeling that it's a reputable site, or it's a business at some place that's where they can actually get what they want, and it's going to give them good customer service and so forth. There are a lot of scams on the web, and there are a lot of low quality sites too. Sometimes you just want to kind of see what people have to say about something. But, other times you want to go to the more authoritative source like the company that actually built this product, and see what the specs are, and see what it costs, and maybe even buy it. And so, that's when it's important for the company to emphasize that they are the official source.
Eric Enge: I also think it's important to try to figure out the user intent. For example, if you have a site that offers digital camera reviews then you want people that are interested in reading reviews. So, you want to make sure that the word "reviews" makes it into the title somewhere. This is because the person is looking for reviews and sees 10 sites about digital cameras, and then your says "digital camera reviews", bang they are on their way over to your site.
Jakob Nielsen: Exactly. It will actually do double duty, because sometimes people actually include the word review in their search if that's what they want. It's such a simple point, but again the users have so many choices, and they can't go to all of them. So, they have to decide, and how they are going to decide where to go?
Well, they make those decisions based on the prediction of the value they are going to get. And, those predictions are based on many criteria, one of them being whether or not you are a famous company they've heard about already. There is nothing much you can do about that, but if you are famous, put your name there. But, the other criteria are things like, does it actually say that these are going to be reviews or tips on how to buy a camera.
Whatever type of information it is, put it there, or if it's a special price, put there that you can get this camera at a vast discount on the price. People are not going to click on a discount of price if they are looking for a review, but the next day when they are looking to buy it, that's when they will click that link.
Eric Enge: Right. And, given that the search engine also puts together a description that it shows, you need to put some thought into the text on the page. You also need the supporting text on the page to reinforce the basic messages to the user. One problem in the search engine world is that sometimes it uses the meta description tag for the description.
Jakob Nielsen: Exactly. And so, it's worthwhile writing a good little short 2 line summary of your pages. This is not always picked up, but sometimes it is, and when it is that's great, because humans can write a much better summary then a computer can. When it's done well it can really be very helpful to the users, and being helpful to the users means some more clicks, and more business.
On the other hand sometimes the search engine decides in whatever way it decides, it is going to make a computerized abstract by pulling together 2 or 3 random little snippets from parts of the page and stick it together, and that's really difficult for users to read.
When that does happen there is nothing much that one can do about that, other than that you can try some of the main queries to see what they put up. If what comes up looks particularly strange then try to reword a few of the words around that little snippet and see if that can make it better.
Eric Enge: Right. Let's talk a little bit about the perception that users have of graphic images. Why do users interpret them as ads?
Jakob Nielsen: It's because people have evolved a very aggressive hatred of advertising on the web, not in the sense of being anti-commercial, anti-business, because people want to buy stuff. They bring their credit card out quite often and are ready to buy things, and so it's not an anti-business feeling, it's more that advertising on the web has become so polluting.
It's information pollution, it's yelling and screaming at people and bouncing up and down and slapping their face, and because the advertising is so aggressive now-a-days, and a lot of pages are filled with blinking, flashing, moving things, people are evolving a protective mechanism to protect themselves from that intrusion on their peace and quiet. The protection mechanism is called selective attention, and selective attention is not something that evolved for the web, it's something that humans have had since the Stone Age, when we were out hunting for the wooly mammoths. If you are out hunting for wooly mammoths, you pay more attention to the trail and the tracks and the signs for where that animal is going. If you are on the web, on the other hand, you mentally screen out the ads, using peripheral vision.
The human visual system is quite interesting, because we think that we have a 180 degree view of the world and we can see a vast amount of things, but, we can only see a tiny little spec in high resolution. This is called the foveal vision, and it is only a small percentage of your eye that has it. That's what you use when you are reading, or when you are trying to identify somebody's face, or when you are looking at the tracks of the wooly mammoths. Any kind of situation where you need to actually see the details is foveal vision, and the rest of your visual field is called peripheral vision. It's very low resolution, very coarse and can only show you a very broad kind of color, shapes and whether or not there is movement.
The ability to see movement is there, because let's say you have a tiger sneaking up on you, you will notice and then you can look at it and then you can see that it's a tiger and you can run and escape and therefore survive.
Eric Enge: In this analogy the graphics are the saber-toothed tiger.
Jakob Nielsen: Exactly, and they are attacking us, but unlike the tiger, the response is it's not something you need to pay attention to. Something that has that general shape might be an ad, and you know that if you look at ad it's just going to be annoying. Therefore it gets ignored.
Eric Enge: So, let's just talk about a corollary to this then, which is a web page that has a set color scheme. Its white background, with black text and this makes it trivial. But, over on the left there is a box which is in yellow. Even though it's text in the box, this may still be perceived as an ad.
Jakob Nielsen: Yes, it will because it has a lot of the characteristics of advertising, which is that it's box shaped, and it's at the edge of the page. It's also has a contrasting color, and contrasting color you might have thought make it stand out, but what actually happens is that people think that it's a foreign element and it's probably an ad, and it ends up being ignored. We see this so many times that websites put up what they think to be a promotion or special feature, something they really want to have people pay attention to, and in fact a lot of the users completely ignore it, because it looks too much like an ad.
Eric Enge: Right. So, the very act of trying to call attention to it, it works, in a sense, because it tells the user: "don't put my high resolution vision on it".
Jakob Nielsen: Exactly, the better way is to put it in a place where people tend to look the most which is the upper middle or upper left of the page. People have a tendency to scan web pages in an F pattern, which is they look across the top, and a little bit less across a little bit further down and they kind of scan down the left hand side of the content area. So, those kinds of 3 places are where they have the most attention, and then you have very little attention towards the periphery of the page, except when they want to do things like find customer service, find contact us, then they know to look at the bottom of the page so, but usually they don't look there.
Eric Enge: Right. So, the right strategy to call attention to something is based on placement, not on flashing lights or colors.
Jakob Nielsen: Exactly, because those mechanisms have been destroyed by all this overly aggressive advertising, and it doesn't even work very well. The best ads on the web are the small, simple text boxes that Google and now many others have as well implemented, and they work partly because they are in fact quite simple and not obnoxious. The other reason they work of course is because they are search driven, so they are exactly what people are looking for.
Eric Enge: Right. Speaking of search I saw a recommendation in one of your posts that search boxes need to be 27 characters wide, as opposed to 28 characters or 26 characters.
Jakob Nielsen: That's right. The reason for that is that 27 is the number that will accommodate 90% of user queries. This is an average across all websites, and so the actual recommendation could be different for specific sites. So, the real recommendation is to do an analysis of your own search logs and see what things people are searching for on your website. The average recommendation across all websites would be 27 characters.
If you are a medical website for example, where the terminology tends to be longer words, very complicated, hard to spell words, then probably it would have to be a wider text box, because you want to accommodate this complex medical vocabulary that the patients or the doctor would have when they are searching. On the other hand, there might be other websites where people are mainly searching for very short and simple things in which case they could have a smaller box, but on average it would be 27 characters.
Eric Enge: Right. The goal is to have roughly 90% of your web queries satisfied without the box having to scroll.
Jakob Nielsen: That's exactly right. You might say why not by 100%, but the reason it's not a 100% is that when you plot the curve of the length of these queries, it the length starts shooting up very drastically after 90%. There are always a few really long queries where people do a cut and paste of some large thing and they paste that into search box. You cannot accommodate these extremely long queries without scrolling, but you should accommodate most of them.
The reason for that is that if the search box is too narrow for people to actually type their query, then the probability of typos goes up tremendously, because people cannot see the entire query anymore and that leads to misspellings, and this leads to people not finding what they want, and lost sales.
The one exception to this general guideline would be for websites where search is the primary thing they do, which is to say if it were a search engine, then I would say make it even wider because then you would want to accommodate 95% of queries. This is in fact what you see when you go to any of the search engines since they have an even wider search box. But, that would just not be a proper recommendation for a business website because search would be a less important thing there, but if you have a specialized search page just for doing search, there you should have a wider search box.
Eric Enge: Right. What you are doing with this recommendation is you are just trying to put it in balance.
Jakob Nielsen: Exactly, all other things are a trade off, and so what I'm saying is 27 characters would be the search box on your homepage, and on your main content pages, because it's always a trade off between doing something else with that space. So, you can't make it as wide as you would, if you were Google, or as you even do yourself on let's say the advanced search page of your site. But, for the main pages 27 characters to start with, and then do your own analysis of your own data to refine it later.
Eric Enge: Right, and, then use the other space you were thinking of using for the search box on something else that's the core of your business. We talked about the whole issue with images being seen as ads, but I've also see some data that suggests that sites that have a certain amount of images and visually appealing stuff score well in usability testing. I also saw your recommendation recently that if you do use graphics then make sure you are showing real content and not just decorations.
Jakob Nielsen: Exactly. I think that's really the main guideline, because it's not really text versus pictures, even though text does have the advantage of being easier to index and search and so forth, but images have another role, which relates to the old saying that "a picture's worth a thousand words", and there are cases where it's true. Let's take a very simple example like, how does this product look? Let's say that it's a fashion website and you are selling dresses, people want to see the dress or they are never going to buy it.
Obviously you want to have photograph and illustrations of the product and from many different angles and close-up, so you could see the details of the stitching and things like that. So, lot of illustrations can be good as long as they are content, meaning that they are information that the customer would be interested in.
They can also be things like a logo, because it provides a feeling of credibility that it's in fact a real company that I know who they are and so forth and gives a sense of consistency and recognition as you move through the site of that it's the same site. There are definite reasons why you want to put things like that there. On the other hand put if it's just smiling models out of a clipart catalogue, that does no good. When we do eye tracking we see people's eyes just move right around them, because they just don't look like they would be worth the attention of an actual fixation, which means to look at them in detail.
That's what I call an obstacle course, some of those images become an obstacle course to a people's actual appreciation of the page, because you have to look and look around so much junk, so much irrelevance to get to what you want.
Eric Enge: That's really the underlying guideline, right, which is if the picture really fits the context of the page, and the expectation of the user, and its content then, yes it belongs there. If it doesn't fit the user expectation then they are going to fall back on it must be an ad.
Jakob Nielsen: Exactly. There are many cases where it's good to have pictures. Another example would be in the PR area of the website particularly for a big company. You should have biographies of the company's leadership like the top management, and this is something that's not just for the journalist, but actually for customers as well. People want to see whether this is a real company, who is running it, and is it sort of somebody that is going to be around tomorrow as well. Having a PR area is important for bigger companies, or even mid size ones too. When you have the biography of the CEO or the Senior Vice President and so forth, we'll put a photo of the person there, so you can see who it is.
There's even a second recommendation too. We've done a lot of studies of journalists where we look at how they use websites to get information for their stories and their newspapers and magazines and so forth. The photo editors want not just a little photo that's on their website, because they can't use that. They want to be able to download a high resolution crisp, great photograph as well. So, have a small photo on the web page, and then have a link to download a really big high resolution photograph, and just that simple little guideline is going to get your picture in a fair number of articles, where maybe the interviews have a lot of people, but they can't get the other guy's photos before a deadline, so they are going to use yours, and you will stand out.
Eric Enge: Your impact in that interview is going to be greater. Well, that makes sense. I have just a couple of questions to wrap up. One article out there I read about your web site said it's the usability guru's website that people love to hate. I'd love to get your comment on that and also how your site scores for usability.
Jakob Nielsen: My personal website is probably not a role model for anybody else except if you are usability guru and you have been at it for a very long time. My website is like a protest against some of these overdone sites, so I have kind of gone overboard in the other direction, but there are some benefits of that. The benefit is that you want to emotionally engage your users, and you want to stand for something, and that's really important to web. There is so much stuff there that unless you stand out and you are kind of remarkable, a word that I think Seth Godin would have used, you might as well not be there at all.
I stand for what you might call even an extreme position on plain spoken ease of use. In any real projects you have to kind of balance, but I don't have to do that, because I can just do what I want to do. So, my site has this very strong positioning which is a little bit overboard possibly, but it does have a lot of benefits, because it is very simple.
It'll work on any computer even if you have a really old computer or if you are blind or otherwise have an accessibility problem, it will work nicely for you. It's very easy to search, it scans, you can make the text big if you are an old person and have a weaker vision. I don't try to do anything fancy so there are a lot of benefits to that.
Eric Enge: Ultimately it doesn't take you long to figure out what you get.
Jakob Nielsen: Exactly. It doesn't look pretty, but I'm not a graphic designer, and I always say to other people to hire a graphic designer for their website because, in general you want to make it look good. Most companies don't want to make the statement that we are about usability unless you happen to be a usability person.
Eric Enge: Right, okay. I learnt recently that Shari Thurow, who is very much into usability as well, is trying to trademark the phrase "search usability". What are your thoughts on that?
Jakob Nielsen: I don't think you can actually trademark that, because it's a very generic term and I certainly know that I haven used it at least back to 1998 and possibly before, and I'm sure many other people have as well. You could possibly trademark it in the context of a certain particular product like "the Search Usability conference". In our company we produce something which is called the User Experience conference, but the term user experience anybody can use.
I don't think that you can take something that's a generic phrase like that and say it's a trademark that only one company can use. I think you can possibly do it for as descriptive part of a bigger name, but then it would be the bigger name that will be trademarked.
Eric Enge: That makes sense to me. Any other last comments that you would like to make on the general topic of usability.
Jakob Nielsen: There are many, many things you can do regarding usability, but there are two main things you can do, and they fall into applying existing knowledge and getting your own new knowledge. Applying existing knowledge: we have done a lot of research already on what makes websites easy to use. We have lots and lots of guidelines that are among other places published on our website. That's the first thing you can do is check your website against these guidelines: the documented best practices that have been known to work and have been tested with thousands of users and thousands of sites already.
The second thing is test with your own customers, because in no way can I or any other usability expert come up with all possible knowledge about your particular industry except of course if we did a custom research project for that industry, but in general you cannot. So, there are always going to be some special issues for any particular company, any particular industry, and they will never be able to discover them just by reading reports or books or websites. They have to do the study themselves, and so do user testing, I just can't repeat this too many times.
Do user testing with your representative customers, even if it's only 5 people. This is enough to find the vast majority of really serious usability problems on your website. I guarantee you, there will be something; every single study always has big findings. There are always so many things wrong that even the smallest study will find a big list of things to change. And then, if you change them the average outcome is a doubling of the conversion rate, and a doubling of the sales of the website.
Eric Enge: It's a reasonable suggestion too, that even after you have been running it a while, test it again.
Jakob Nielsen: Yes, you are never done. You can always get better. I have of tracked these things over time, and what happens is that as the science gets better, the user's expectations actually get higher, so people want more and that might be unfair, but that's how it is. People want more and you are up against competing sites that get better as well. So, yes, you are never done, you continuously have to improve which is a basic quality engineering concept: Continuous quality improvement and that definitely applies to usability as well. Keep testing and you'll keep discovering interesting additional insights and changes in user behavior and changes in what they want and so forth. But, for sure if you have never done one before, just do it now, and then maybe do another one next year, but definitely do one now.
Eric Enge: Right. Well great, I have enjoyed the conversation.
Jakob Nielsen: Yes, thank you.
Other Recent Interviews
- Ask.com's Gary Price - Sep. 10, 2007
- Incisive's Kevin Ryan - Sep. 4, 2007
- Market Motive's John Marshall - August 15, 2007
- Danny Sullivan - August 13, 2007
- Clickz's Rebecca Lieb - August 6, 2007
- Robert Scoble - July 23, 2007
About the Author
Eric Enge is the Founder and President of Stone Temple Consulting (STC). STC offers Internet marketing optimization services, including SEO, Social Media and PPC optimization, and its web site can be found at: http://www.stonetemple.com.