In this interview, usability guru Jakob Nielsen takes aim at RSS, Flash and the design failings of the consumer electronics industry.
He calls Google’s non-search products a “hotch-potch of weird stuff”. He digs into sex and segmentation. And he also reveals his day rate…
What problems are new technologies like Ajax creating for usability experts?
I think the main problem is that they divert attention away from the most important things – things like clear communication, good product photographs, and simple navigation and search.
Ajax allows some things to happen a bit faster, but for the average website that isn’t the question. Users aren’t typically on the same page more than once anyway. There are cases where people do repeated actions, for example on a stock trading website where people manipulate listings for their portfolio. That could be a case where increasing the efficiency of these actions is important.
But on an e-commerce site where people are browsing through it, the question is not whether something is two seconds faster, it’s whether people can find something or have to spend five minutes doing it. Ajax can provide improvements in one area, but most problems on websites are of a vastly bigger magnitude.
You’ve talked a lot about participation inequality. Do you think any of the big community sites are tackling that effectively?
Probably no. And I don’t think it’s something that you can completely overcome. There are some sites that have taken steps to give enhanced profiles to more active participants. That encourages people to give more, and it doesn’t cost you anything to give them something back as what you often give them is free.
But that said, participation inequality will always be with us. It’s about shifting that graph so that 50% of users, rather than 90%, don’t participate. There will always be a large amount of people that don’t bother.
Do you think there’s an argument for a more segmented approach to usability from designers?
Mainly not. The main point is the gap between the company and its customers – not between different types of customers.
The first step is to find out what your customers want, and then to go on and find out what different customers want. That, however, often tends to be a paralysing event – you can do the profiling so narrowly that almost everyone ends up having their own profile, and then you’re talking about personalisation which is a different question.
There are sites where your customers are so different that you would want to treat them differently. A good example would be a trading site like eBay where there are buyers and sellers.
But there are a lot of sites that start off with a question about whether you are a small business or a big business. That’s often annoying for users as small businesses sometimes need a heavy duty solution and big businesses sometimes need something for a small internal group.
What about gender? Are there big differences between what men and women want?
Yes, but funnily enough, not much. For most things, we’ve seen differences of around 1% between men and women’s behaviour, but there are gaps of 100% between how computers and companies internally think about some things and how customers think about them.
That’s in terms of behaviour – in terms of content, there are obvious differences, but in terms of how you would do your navigation or photographs, those things are the same.
It’s still quite a male-dominated industry though. Has enough research been done into how sites should be designed for women?
That’s true, but I think the usability field has more women in it than men, for whatever reason. We have a majority of women in our company. More importantly though, in our studies, we always try and get an equal percentage of women in.
The only big differences we see between users are age differences. Senior citizens are different from the mainstream, which we see as people from 18-65. Children are very different from teenagers and grown-ups and old people.
Do you reckon any of the traditional rules about usability need updating? The eight second rule for page loading, for example, with the rise of higher speed connections?
Well, the real rule is actually one second. There are three rules for response times and they don’t really change as they are based on fundamental human characteristics.
The rules are; if it is faster than one tenth of a second, you don’t feel like you are waiting at all. If it is more than one tenth of a second, you can tell you are waiting, but up to one second, it still feels like smooth navigation. Between one and ten seconds is the limit for your attention.
As you wait from one to ten seconds, your attention starts drifting off, and after ten seconds you are asking ‘where is this thing?’. The recommendation is that it doesn’t need to be one tenth of a second, but it should be faster than one second as it’s about free flowing navigation. The only reason we used the eight second rule is that we couldn’t get down to one second in the past.
If you go the best websites, like Google, that’s what they do – they give you the page like that [clicks fingers].
What about other rules?
Well there is one where I’ve changed my recommendation. It’s a very specific one but it is how you show hypertext links. I used to say they should be blue, but now it is they should be coloured. That’s because in the early days of the web, blue symbolised ‘click here’. But now any colour symbolises ‘click here’. That’s an example of where things can change because of people’s experiences.
How has that area of design been affected by text link ads?
They add to what people have seen, and people really go by what they have seen.
One of the over-reaching usability guidelines is to go by what people already know. Don’t try and do something new on your website, because you are not important enough. People think they are exceptional, but no website really stands out so much that people are willing to learn different behaviour just for that site.
Web use is about flitting between sites and spending a few minutes on each site. Even in the B2B space, where there are extra-specialist users researching complicated products, they will tend to visit four or five competing sites and compare the products. No one site has a monopoly on people’s attention – it’s not like the physical world.
Are you surprised by the amount of pop-ups that still exist?
Yes, I am actually. But at least the good news is that users are seeing fewer and fewer as pop-up blockers are becoming more common. That’s a great piece of evidence for my point that when something becomes so annoying the defence mechanism starts to become very popular. And that’s bad.
What would you say to a publisher that still offers pop-ups to its advertisers?
I would say wean yourself off them. You will have to soon anyway, because everyone will soon be using a browser with pop-up blockers and nobody will see them.
But also, advertising revenue isn’t everything – your brand loyalty is valuable as well. We know that people don’t just ascribe the annoyance created by pop-ups to the advertiser but also to the publisher. People have a negative reaction to the publisher if the ad is hosted on that site. The site is blamed for it.
What do you reckon is the best format for video ads from the perspective of usability?
Mainly short videos, because that fits in with the surfing experience. People spend a few minutes on each site and an even shorter time on each video.
This is how TV advertising tends to work as well – it may be that TV ads work well for online video as they are heavily cut. I think long videos are not good for the web, but I want to distinguish between the internet and the web here. Through the internet being a distribution mechanism for TV programming, there will be video on demand and you can download a two-hour-long video and put it on your TV through wireless networking or something.
That’s internet TV, but it’s different for computers when you’re clicking. That’s where short videos will work.
Are there any products and services away from the net that you think are crying out for some investment in usability?
I definitely think so. Almost all of them, to be frank. The worst offenders, I think, are consumer electronics. There’s a hotch-potch of incompatible equipment. You have all these wires and the average person doesn’t know how to connect them as they have ten outlets. That is very poorly done. You typically have five to ten remote controls you need to watch a movie. That is very, very bad usability.
That’s probably the number one thing. Mobile phones are also still at the point where they could use a good usability boost.
Are you getting much work from either of those sectors?
Very little. Certainly very little from the entire consumer electronics industry. As a company, we are busy enough doing web and intranet and computer projects but it’s sad that this whole industry is ignoring their user base in a sense.
They don’t tend to care about usability. They are in a kind of competition over features – it’s where the software industry was ten or fifteen years ago. But people don’t care about features if they can’t use them. Microsoft was doing Microsoft Office and they did a survey with some of their big customers about what features they wanted in the next version. They got a lot of requests, but it turns out almost all of the requests were for features that were already in Office. People just couldn’t find them.
In the version that came out just a few months ago, that was their biggest push – making existing features easier to find, rather than adding more.
Have you been asked to do any weird or interesting projects recently?
Well, we had one where we worked on elevator control buttons. It wasn’t something that was crying out for usability, but it’s an example of the fact that anything can be made a little easier to use if you bother to look at it with the right mindset.
What did you recommend?
Well, that’s confidential from the point of view of the client. I wouldn’t say we made these elevators twice as easy to use, which we typically say about a website. But if you think about the amount of time people press an elevator button a day, even a few percent better would be worth doing.
What do you think has been holding back RSS uptake?
First of all, the name. I think it should be called newsfeeds, as that is something that the user has a chance of understanding. It has a new logo that’s much better though – the old XML logo was even worse than RSS; really bad.
But the main point is that there is a vast group of people that don’t necessarily want that incessant amount of updates. One of the things we found in our studies of email newsletters is that people have different levels of tolerance for information flows. Some people just like a monthly, weekly or daily newsletter, while some people want newsfeeds in real-time.
There’s a whole range there, but there is a tendency for what you might call ‘the internet elite’ to be among the most active users and are therefore very keen for the real time updates and having all the information right there. Then there are other people, like normal businesspeople or senior managers, who are really busy and just want a monthly newsletter.
One guideline is that if you do have a newsletter, offer it with several different frequencies. If people unsubscribe, say ‘sorry to see you go, but we have a monthly newsletter if you want that instead.’ You may keep some of them.
Do you think RSS readers have limited potential growth then?
I think they have an inherent limit just because they are targeting the most active users. That goes back to your question about profiling or segmentation – that’s an area where it does make sense to have segmentation. It makes sense to offer different frequencies – from monthly newsletters to hourly updates.
Is there any point in using Flash on your website?
That’s a good question – I’ve been going back and forth on that over the years.
We’ve come up with examples in our studies where it has worked well – product demos, for example. There are cases where you can enhance the physicality of a product by showing a little demo. Also there are some examples on health sites where you can persuade people to take their pills. That’s an example where animations can show complicated things simply.
But that said, 99% of the time it comes across as completely stupid and annoying.
Do you think there’s more acceptance now of the importance of usability at board level within companies?
I definitely think so. The acceptance definitely has gone up ever since the dotcom bubble burst. That was the defining event that showed senior executives that it wasn’t just magic and you have to do what users want.
Also, the success of certain types of web companies – the ones that have been really successful – are all the sites that are very straightforward; Google is the most famous one, but Yahoo! too, as well as Amazon, eBay, Wikipedia and Craigslist. That point has gotten through now.
Is Google still the best site for usability on the web?
It probably is actually, but on the other hand, one of the reasons it is so good is it has a very clear mission.
Google’s not so good if you go beyond that mission. They have a lot of services that are really a hotch-potch of weird stuff that’s just kind of lumped together. The totality of Google is not necessarily the greatest way of offering online services, but the search service completely is, and of course that’s where they make all their money.
What should Google do to solve that?
They should have a more integrated approach. That’s where I think Yahoo! has done a much better job – having a lot of different services that are nicely tied together. You have a feeling of a rich service when you go to Yahoo!
What’s your day rate at the moment?
Well, it tends to be a bit on the fairly high side. Sort of ten or twenty thousand dollars.
What’s happening to usability costs?
It depends what you are looking for. In the early days, there was almost no-one that knew anything about it, and I was one of the few. So at that point, you were hard pressed if you wanted this advice. Today, on the other hand, there are a lot.
It’s kind of a pyramid where there is a tiny amount of people at the top that have twenty years’ experience, and there’s a very large bottom part where people have one year’s experience. There are more and more of them and to be honest, a lot of products don’t require a world expert to work on them.
So it’s key for a client to work out exactly what level of experience they need to buy in?
Exactly. There are many things where I wouldn’t say anyone could do it, but you don’t need a lot of experience. And there are a lot of things where you would benefit from going to the more senior people.
But that’s the same with anything – sales, marketing, accounting or HR, any field.