       | ||
|
:: back :: |
|
:: articles :: - www.nmk.co.uk - Jakob Nielsen on Homepage Usability By Colin Kirkpatrick NMK: Your new book is called Homepage Usability. Is designing a homepage all about creating a good first impression? Jakob Nielsen: It would say it's partly about creating a first impression, but it's also about creating a good user experience for repeat visitors. The most important role of a homepage is to communicate what the company is, but a lot of websites don't do this well at all. For example, you should include a simple tag line that gives a clear definition of what the company is actually about, but this is actually quite rare, and technology companies and start-ups are often the worst offenders. Also, you need to show what the website is about - what can the user do there? So that one simple thing is really important. The other thing is get out of the user's way in terms of 'fluff'. One of the things we do in our book is breakdown the screen real estate of the websites to show what percentage is actually of interest to the users, how much is only of interest to the company and how much is just meaningless filler. NMK: Does the use of jargon contribute to this problem? JN: Yes, completely. It's also about graphics and other design mistakes, but it's amazing how the language of the web is not as it should be, it's not concise and to the point. The vocabulary has no communicative value. You could shuffle the words around and change the labels and it wouldn't be any more or less meaningful, and that really indicates to me that it doesn't work. NMK: How did you and co-author Marie Tahir choose the 50 sites that are deconstructed in the book? JN: Most of them are selected from various top ten lists in the US and the UK, and then we added a few from other places like Australia and places in Asia where they had English language websites, and also some government sites and small company sites. We tried to have a broad diversity, that was the basic goal. Another goal was to choose sites that are basically good. If we had deliberately gone out to look for the 50 worst homepages in the world, that would have been really horrible! So these are all from respectable places-but we still came up with more than 1000 comments about specific usability issues. NMK: Of the 50 websites, are there any that really stand out? JN: One that is particularly weird is for an Australian supermarket called Coles, which is quite a non-standard website. First of all, its one of those sites that's devoid of content, and there's nothing to say that this is one of Australia's biggest supermarkets. The tagline says 'serving you better' but they would never say 'serving you worse' so this really doesn't have much meaning whatsoever. And they actually play muzak! One of the benefits you get from shopping for groceries online is that you don't have to listen to the stupid music, but they've taken one of the bad things about shopping at the supermarket and put it on the website. Another interesting point about this website is that it violates one of the fundamental layout rules, which is that you put your logo in the upper left-hand corner of the page. In 84 percent of cases, the company logo is in the upper left-hand side of the website, so this is where users will look first. It's a very good example of what I call [Jakob's Law of the Internet User Experience], which is that you spend most of your time on other websites, therefore you base your expectations about how the web should work on your accumulated experience of all websites. It's a very hard law for a lot of web designers to obey, because you have to be humble to follow it, and accept that the internet is vast and you are small. To the users, the web is the world and you are just a tiny bit of it, and that's really important when doing the design. NMK: Is the web improving in terms of usability? JN: I think it is, and if we'd done a book like this five years ago it would have been much worse, particularly for big-company websites. Five years ago they were generally nothing but a scan of the annual report -sometimes literally- and that has improved. Web usability used to be dominated by blunders, incredibly stupid design mistakes, like the big flash intro or huge pictures that take half-a-minute to download, and it only takes a very small usability study to see that it's wrong. It's taken us about seven years, but I would say that usability has now mainly stamped-out completely user-hostile, stupid design. We're now at the next step. Just getting rid of the idiotic mistakes that make a website horrible to use doesn't mean that you automatically achieve the opposite, which is a wonderful website that supports everything users want to do. That's why we now have more detailed guidelines - the book has 113 guidelines on homepages, whereas in the first seven years there was essentially one guideline, which was 'don't do stupid things that get in the way of users'. Now we can say, 'okay, so do some things for users’. NMK: But aren't a lot of these guidelines just common sense? JN: Yes, but it's taken seven years of fighting for people to accept common sense. I think the problem was that it was a new medium, and a lot of people were still trying to use their insights from earlier media, and that's understandable. But it's still inexcusable, because all they would need to have done is take five of their own customers and watch them use the website, and there's no excuse for not doing that. Even if you don't believe me you can just do your own study. NMK: Can small companies with limited resources do their own user-testing effectively? JN: The point is that you can do a large number of things on a small budget. Everyone can get the guidelines and apply them to their own design, that is very cheap. It costs nothing to get the guidelines from my website (www.useit.com). The next stage is to run a test with your own customers, and the happy fact is that you don't need to have a huge bunch of them. If you're designing a new cereal box, you're got to test it far more extensively. But the cereal box was designed 50 years ago and we have a perfect cereal box already. You're just tweaking it in minute detail, trying to make a one percent improvement. Now on the web this is not the case at all. Because web design is in such a crude state, we are looking at making design changes that will make a 100 percent difference, that will double the use of the website. These improvements are pretty easy to find if only you bother looking for them. The methodology is really simple, you need only a handful of users -five is a good number- and the entire process can be completed in three days. You sit next to the user and watch what they're doing, set them tasks, see where they fail, find out why they click the wrong button. And you need to know what they say at the moment they click on it, because you don't want indirect, retrospective thoughts about it. That's why I say don't run focus groups if the goal is to improve an interactive design like a website - you've got to see people use it, not talk about it. Also you have to look at one person at a time, not five people in a group, because everybody will use it differently. That's a very important distinction. The key point is what they actually do; it's a matter of facts, not opinions. You do this with every different design you have, or you do iterative design, which means you test it and you change it and you improve it and you test again. The more small tests you can do the better, that's the starting point. NMK: What is the next big issue in usability? JN: The big issue is to change usability from being a design-critic to a design-driver. We are still in the phase now where usability is used to improve bad design, but the next level is you don't do the design at all until you've done usability research to identify the users' needs. You should go and study your users before you do the design. NMK: Some people within the design community think you are in favour of utilitarian, creatively boring websites. Are you spoiling our fun? JN: I think that utilitarian is not boring I think it's exciting, and I think it does not stifle creativity whatsoever. On the contrary, it unleashes creativity once you start thinking about how we can support people better. I think the problem is that you have certain groups of designers who think of web design as being an art project, and its not. If you want to design art, go and paint a picture and put it in a gallery and I'll look at it there. If you've designing something that's a tool, you've got to design something that works, because people don't go to a website to sit and admire it, they go there for a reason. These days the good designers really view usability as being their friend, because it's a way to spark creativity, just in more constructive ways. If you think about it from the users' perspective, having something that works is not boring, it is exciting. What is exciting about the web is when it allows you to do something. That was one of the biggest lessons we learned from our accessibility study (Beyond ALT Text: Making the Web Easy to Use for Users with Disabilities): in spite of all the many problems documented in the report, the bottom line was that all these people with disabilities said 'the internet is liberating, it is empowering, it allows me to do things I couldn't do before'. And that is exciting. NMK: What do you see as the key usability issues for users with disabilities, such as visually impaired people? JN: The first thing to realise is that there are special concerns, as a lot of websites never even bothered to think about it. Then I think the next one is that you've got to think about users with disabilities as being users, as people with tasks and goals they want to perform. The mainstream approach to accessibility is actually quite misguided, because even the people who do think about it often have the wrong approach, which is 'lets just put all our stuff online in a text-only format, or in a way which is theoretically accessible'. But I really want to make a distinction between accessibility and usability, because usability means the person's ability to use it, and accessibility means that they can theoretically get at it if they try hard enough. We know from our studies of sighted users that just because all the information is there on the screen doesn't mean that people understand it, or can use it, or can find what they're looking for. It's completely the same when you look at people who are visually impaired, for example, because they have usability issues just as sighted people do, it's just they have a different interface, which is a voice interface. So if you take a web page that is theoretically 100 percent accessible, you may still have a miserable interface from a usability perspective. For example, if you have really long menus, sighted users can scan the menus and pick out where you want to go, but if you have to listen to the entire menu using a screen-reader, it will be much harder to pick the right choice. In our study (Beyond ALT Text: Making the Web Easy to Use for Users with Disabilities) we found that it was about three times easier for sighted users to use the web than visually impaired users. Even though they could do it in theory, in practice they often could not, because it wasn't designed to be easy. NMK: The British government wants to have 100 percent of it's services online by 2005. Are there any lessons it can learn from e-government sites in the US? JN: It sounds good to get everything online by 2005. But it could be the wrong way of going about it, because it's an internally motivated project. What you really should do is say 'we want to have X percent of all customer transactions online'. First they should find out what the users are looking for, and then they should make that as easy as possible. If they just put everything online they might discover that they aren't helping anybody, because it's in such a format that nobody can find anything, or it might be so technically sophisticated that only the most technically sophisticated ten percent of the population can actually use it. And of course the government is a special case because it has a responsibility to everybody, and that's a much harder usability goal. NMK: Has the economic downturn taught us some harsh lessons about what we should and should not be doing on the web? JN: I think actually it's healthy because there was so much excess in web design and there were so many web services being offered that didn't make any sense, and they were artificially pumped up by this influx of money that didn't know where it should go. There was so much money sloshing around that the most stupid idea could get funding - or at least the second most stupid idea. I remember cases where a website in the States went public for a very large amount of money, something in the region of a billion dollars, and the analysts said it was worth this much because they had so many unique users. I calculated this meant that each user was worth 10,000 dollars! A customer is only worth 10,000 dollars if they do 10,000 dollars worth of business with you, and that's obviously not going to happen just because people go and look at your homepage. NMK: Are metrics such as 'unique users' or 'return visitors' ever a reliable way of measuring the effectiveness of your website? JN: 'Return visitors' is a reasonably good indicator, but it depends on what your goal is, so if your goal is to value your business you have to look at what these customers actually do. If they don't do any business they have a value very close to zero because we now know that regular advertising doesn't work on the web, so they've got to do some actual business. Now if it's not an advertising-driven website but rather a website you are doing for your company then ultimately it's still the same, it's how much business do you get? But quite often the business would come not from people buying on the website, but from people buying in the stores, or through the sales force, or the distributors, and therefore it's more indirect and harder to measure. That's where I would still place some value on metrics like 'return visitors', but I would always look at users who return to your site, because I don't think any value should be granted to someone who takes one look at your homepage and then leaves. That's a unique visitor, but it's not a user, it's not a customer. I would also look at problems solved. A lot of websites do customer support, but that's something that's often done very poorly. So you look at your solution rate, and that's something you can measure in the lab. NMK: Is it true that you think email on the verge of breaking point? JN: I really do, I think email is on the verge of turning from a productivity tool into a productivity killer. You can spend all day now doing your email and getting no work done. NMK: What technological advances are likely to make the most difference to usability in the next five to ten years? JN: I think we'll see hardware improvements, such as bigger screens, clearer screens. That will happen because they already exist, but they need to become cheaper. We're going to see much better mobile devices, I really think that could happen in 2002, because there's so many devices that are just about to come out. But I'm much more pessimistic about software advances, because I think Microsoft is going to keep pumping Microsoft office and Microsoft windows for the next five years. Maybe after that we will get true software innovation, which I think we need because we are now dealing with so much more information than we were in the past, and the level of support that you have in those office automation tools is just not sufficient for the internet era. NMK: In the quest for common standards, is it desirable for software products and websites to adopt the appearance and functionality of Microsoft products? JN: Well, in the short term it probably is, because there are some benefits to leveraging familiarity. At the same time, I would also say it's very dangerous, because if the look doesn't match the feel then you have a conflict, and that's not good. In the long term it's not a good thing, because it's kind of a limited and clunky user interface. NMK: Are you optimistic about the future of mobile devices? JN: For the future, yes. For the present, no. Last year we did a project on Wap, and the basic conclusion was that it had such a bad user experience, so incredibly reduced usability, that it could not be recommended for anything but the most critical applications, because the user experience is truly horrible. At that time we got a lot of interest from the Wap industry, because these billion dollar companies were trying to push the technology, and we basically came out like the little boy in the Emperor's New Clothes, and said "look they have no clothes on, Wap has no clothes". We were heavily criticised for this report, but by looking at users and their behaviour you could see it was the truth. So this year I tried it again, and it's just as bad as last year - there's been no progress in Wap for a year, it's still as slow, and they still make the same design mistakes we documented in the report. I'll give you one example: I tried to look-up what was happening in the stock market on Vodafone, but first I had to go through three splash screens before I could get anything - and it's so slow that you really don't want to download their logo every time you open the Wap browser. Finally I get to a menu that's called 'financial indices', so I click that, and again I wait for ever, and I get to a screen that says FTSE 100 Index, Dow Jones index and so on. But what they should have done is provide me with a number right there, they should say 'FTSE: 5300', 'Dow Jones: 9800', and give me the numbers right there. They don't have to spend the space on the word "index", because I just clicked on a menu that says "indices", so I know I'm going to get lots of indices. Spend that very small precious space on the phone screen to give me what I want instead of requiring one more step, and that is such a simple principle. copyright © 2002 ebiznet99.com |