In 2006 WUP developed a range of personas for the Environment Agency. These won an Intranet Innovation Award for “supporting the delivery of better intranet content and functionality” and bringing together information that “lacked focus and coherence”.

The Environment Agency wanted to do additional work to gain greater insight and understanding the goals, motivations and contexts of two of the personas – one for the intranet and one for the internet. We proposed a range of research approaches to address these issues including two less usual methodologies: contextual inquiry and an online user research forum.

The contextual inquiry allows an exploration of the work environment and contexts of participants using ethnographic techniques to understand the more ‘tacit’ aspects of an individual’s work place and behaviours; this will be combined with depth interviewing/participative research techniques to establish user needs and attitudes to the sites, and explore possible solutions to perceived problems.

The online user research forum enables participants to become part of a closed, invitation-only group of users who interact with each other. This allows us to gain greater insight into their needs and give greater ‘reach’ than more traditional approaches, and can address small, discrete issues on an iterative basis.

This project is due for completion during the summer of 2008.

When we enquired why this image was on the home page we were informed that two of the directors had dogs and they liked to see them on the home page. Indeed, there seemed to be some competition between them to see whose dog was featured!

Many of the public sector sites we work on are particularly bad in this regard – except rather than dogs they often put their pet initiative - which is seldom of interest to users - on the home page! The problem is that the site owners know precisely what their department does, and because they don’t have a user focus fail to realise that this may not be obvious to site users. Unfortunately it is hard for me to name names and still keep these organisations as clients.

But I can illustrate the point by looking at sites that are not our clients. Have a look at the Ofcom site. I am sure there are lots of different target audiences for this site but clearly the general public is one. Would you think, from the home page, you were in the right place to make a complaint about a TV or radio programme, or to complain about the service you have from your fixed line, mobile provider or ISP? There is a hint you might be in some of the text on the home page, “Ofcom is the independent regulator and competition authority for the UK communications industries, with responsibilities across television, radio, telecommunications and wireless communications services” but this is so wordy that goal orientated users are unlikely to read it – also it’s apparently in the ‘News’ section. There is a further hint in the left hand navigation with the link ‘How to complain’ but about what? The same goes for the ‘Advice for Consumers’ link – advice about what?

In both these examples the site owners are guilty of assuming users know what they do, they’re not a Tesco or a Ford. Some people might know what they do but many more won’t, so the goals users can achieve on the site need to be made explicitly clear with words and pictures so users can immediately understand what the site’s about.

But in many ways understanding the user experience is the easy bit. We all go off, do our thing and present the client with these great insights about users – and we all, including the client, know these are genuinely useful insights. The problem is that the client then has to do something with these insights, but they often seem poorly equipped to do this. I wonder what proportion of our collective insights are ever acted on? Probably less than half – if that!

The more difficult part of making a website effective is to work out what to do with the user experience insights. The user surveys, analytics or usability study may identify the problem but they don’t necessarily tell the client what to do to make things better. This is where expertise on implementation of solutions to problems is required. The client often has little experience outside their own website, but the agency, on the other hand, who works with a range of clients and sees many different solutions, should be well placed to help with problem solving. But how often does this happen? Are we content with diagnosing the problem and letting the client find the solution?

Often even more challenging than identifying the solution is getting it implemented. Organisational constraints such as lack of a clear site strategy, inadequate user focus, inappropriately trained and resourced content authors and lack of senior management support are the major stumbling blocks – much more so than technical issues. How good are we at helping clients address these issues? WUP offers a number of services to help clients with these problems, but it is not easy to get the key people to understand why these are important, and it can take a long time to change an organisation to address these issues.

So, it’s much easier to just provide some user experience insights. But on their own will these lead to more effective web sites?…… I doubt it.

Our mix of work continues to develop. Usability testing is still at the core of what we do but we have also undertaken information architecture and wireframe development for a number of clients, and run numerous strategy and training workshops for clients in order to improve their understanding of their users, users’ goals and how to develop effective websites.

Redeveloped sites launched this year following work undertaken by WUP include: Biotechnology and Biological Sciences Research Council (BBSRC), easycruise, Sock Shop, National Database of Accredited Qualifications, and the British Museum.

WUP’s managing director, Peter Collins, commented:

“It is very satisfying that our services are in increasing demand and that we have managed to attract many major new clients. But it’s even more satisfying that 50% of our turnover comes from existing clients who have returned to us for new work. This clearly proves they are happy with the outcomes of our work and the service they’re getting”

‘I think it is very important to make a distinction between designing a website so a blind person’s screen-reader can reach the contents but also the ability of a disabled person to visit a site, find the information they’re looking for and complete a task at the same time, the same convenience, the same cost as somebody who’s not disabled’

In many ways I couldn’t agree with her more. We have tested many local authority sites with able bodied users, blind users and users with other disabilities. And the sad truth is that most sites are not very good for any of these users. In our tests, able body user goal success rates tend to be at about the 50-60% rate. However, disabled user goal success rates are not that much worse.

If a site can cope reasonably well with the disabled user’s access technology (not all can), then the problems disabled users face are very similar to the problems faced by able bodied users: things like poor information architecture, poor page layouts, too many links on a page, ineffective searches, and far too many words, words, words! These are usability issues that make the sites bad for everyone!

What the disabled users do is throw a really strong light on just how bad these usability issues are. If you are a blind user and arriving on a web page and Jaws (a screen reader) has just intoned ‘This page has 142 links’ (check out the home page of Birmingham City Council!) your heart sinks. But the able bodied person has the same problem; it may be quicker for them to scan the links but there are still too many of them. The blind user is forced to listen to the vast amounts of irrelevant words that fill so many Local Authority web sites. The able bodied user can scan over these but it still makes it harder to find what you want. Pages like this need simplifying to make it usable for all users.

A lot of focus is put on making Local Authority sites accessible to disabled people - and so there should be. But there is still plenty of work to be done to make them accessible to able bodied people as well! I think Julie’s bench mark of performance for disabled users is wrong. They shouldn’t just be aiming to make them as easy to use as for somebody who’s not disabled. We should be aiming higher than that. They should be making them easy for everyone to use.

However, there are times when users seem to easily find information that is located on the right of the page. This set me thinking about when this part of the page can be used with confidence that users will notice what is there.

Various authors have reported that users look at pages in an F shaped manner. Our own observations confirm that generally this is the case on Home pages (though not always) but that this is not normally the case on navigation pages (pages that lead a user to the next level in site hierarchy) or on content pages. What we observe is that, on these pages, users look first in the top centre - but below any banner or navigation - this is where they expect to find relevant content, or links that will lead them to content.

If they can’t find what they want here, they will then look around the page, initially by scrolling down and then, if this is not successful, they will look to the left for a left hand navigation, or at the top. On many occasions users will still completely miss relevant links on the right of the page.

However, if a user is on a relevant content page and has found what they want (i.e. they will have achieved their goal) they may then think of something else they want to find then they will look around in a much more systematic way and will include the right hand side of the page in this search. We recently tested a prototype wire frame for a client where links to follow-on tasks were put on the right. These were found by all users who stayed on the content page to complete the task. (Over half the users in this test did this, but the remainder clicked the back button to start the follow-on task from a higher level in the information architecture).

The key conclusion seems to be that whether the user will look at the right of the page depends where the user is in their journey. On route to a goal, users look mainly in the centre of the page, but having completed a goal they will happily look on the right for related information. This clearly has implications for the type of content that can be placed on the right of a page. On content pages this space can be used for links to related content but on navigation pages it is likely to be missed.

What do users want from intranets? Gerry McGovern’s research on users’ needs from intranets has concluded that ‘Staff overwhelmingly want a better organized intranet where they can quickly find people, policies and procedures, and forms’. His research has shown that there are a limited number of tasks that are really important to users: when respondents were asked to identify their top 5 tasks (out of 58), ‘finding people’ (i.e. a workable contacts directory) emerged as the most important task, and ‘the top 5 percent of tasks (finding people, procedures/processes, and forms) received 33 percent of the vote’.

So, why don’t intranets deliver what’s needed? There are probably 3 key reasons why many intranets don’t deliver:

• Lack of clear site aims: originally most intranets were set up because the technology was there, and it was the thing to do because other people were doing it. There was often inadequate thought about what the aims were for the organisation and the individual users. At a top level, intranets should be about increasing organisation effectiveness (how well it does things) and efficiency (how quickly it does things). Defining what effectiveness and efficiency means in an organisation will provide clarity about why the intranet exists and help shape its key features
• Lack of understanding users’ priority goals: you’d think that understanding priority user goals would be easy for an intranet - after all isn’t everyone a user? And Gerry McGovern’s research suggests that there’s actually a small number of very important things for users that need to be highly visible and accessible - if you just asked users in the organisation “what will help me do the ‘right’ job in the most efficient way”. But it obviously isn’t that easy - mainly because over time intranets seem to become corpulent organic bodies that lead lives of there own - so bloated that the key things that users really want are obscured. And that leads onto the third point…
• Intranets are the ultimate ‘vanity case’ for content authors to publish their wares - whether or not the content is critical to users. The intranet is a good place to vanity publish what you do and justify your place in the organisation. Without a clear publishing model and a robust approach to implementing that model, unnecessary content will emerge and clog up the intranet reducing its usefulness to users.

A small proportion of tasks on an intranet represent the biggest proportion of user goals - so getting users quickly and easily to these ‘most important goals’ will both increase user satisfaction with the site and increase operational efficiency and effectiveness.This requires absolute clarity from the organisation about what the intranet is for, what users want from the intranet, and a ruthless approach to managing content.

What is a trigger word? It is a word or words associated with a link to make it clearer to users the type of information they can find beneath a link. Gerry McGovern calls them care words, other people call them trigger words. Go to the BBSRC site and you will see that under the link ‘Our Organisation’ they have the words: Structures, Policies, Institutes and Centres, Spending, Achievements, and Jobs. These are the trigger words.

Why are the needed? Well the amount of scent that the average link, of one or two words, in the main navigation of the site can give off is limited. As work by UIE shows the best performing links contained 7-12 words - difficult to fit that many onto a tab. Therefore, by adding trigger words - based on user research identifying the words and language that are meaningful to users - you can get the link to give off a lot more scent for a wider range of goals.

For example, you will notice on the BBSRC site that the trigger words are not links, we think making them links is a bad idea. We know it may add another click to the user journey but we don’t think that matters. If they are made into links users will go straight to the content page and miss the navigation page at the next level down. While this may be good for some users, for most users it is better for them to see all of the options in the navigation page because they are then more likely to get to the correct information for them.

User satisfaction is more strongly correlated with the confidence a user has they are going in the right direction than in the number of clicks it takes to achieve a goal. So as long as the link plus trigger words gives of enough scent then the extra click doesn’t matter.

However, something else that came up recently in some user testing is that you should not have too many trigger words. If you do, users may think these are all the goals they can achieve under this link, rather than simply being a way to demonstrate the types of goals that can be got to via the link.

So if your information architecture isn’t working get thinking about those trigger words - users love ‘em!

So, how do they behave? Instead of selecting quickly the first link that looked like it would take them to relevant information, they carefully reviewed all the links on the page to decide which to choose. In other words they exhibit ‘optimising’ not ’satisfying’ behaviour. This is the opposite of what we see when someone browses using a PC. Here, users will often select a ‘wrong’ link because it is the first one they come across that gives off some ’scent’ that it will lead them to the information they are looking for. We have even observed users ignoring good links - ones that give off lots of scent - right below a less good one because they have not bothered to look beyond the first one.

Why do they behave like this? This behaviour is reminiscent of what we used to see 5 or more years ago when users browsed on PCs using dial up connections. Broadband has changed the way we do things. Watching the testers, quickly gave us an explanation for this behaviour. Even though we were browsing using a 3G connection, download speeds were not fast - this was not a mobile optimised site, but more importantly the users were paying by the byte. They were aware that downloading the wrong information would cost them time and money.

I, therefore, wonder how quickly the promised revolution in mobile web use will happen. Clearly, devices like the Apple iPhone have the potential to overcome these problems - as long as you live and work in an area with good wireless coverage - but, unless downloading data on non-Apple devices has a zero marginal cost I suspect users will still be very careful about the links they select.

The client felt the card sorting exercise was of limited value and was not inclined to pay to repeat the exercise. When we looked at the results of the card sorting work we were struck by the fact it seemed to throw little light on how a new information architecture might be structured. This led us to reflect on our own approach.

So, what’s the issue with card sorting? Anyone who has undertaken a card sorting exercise will know that what it often provides is a BGO - ‘blinding glimpse of the obvious’. All the user goals you would expect to go together are put together by users. Goals relating to an organisation’s structure, staff and policies are put together in a group and, surprise, surprise, when users are asked to suggest a label for this group they suggest ‘About us’!

It is the large range of other goals that we, the information architect, have difficulty deciding how they fit together that users also have trouble sorting. Often each user comes up with their own very different grouping.

A further problem with card sorting is deciding what to write on the cards. Users seem highly influenced by the use of individual words. So the repetition of a single word in very different goals (e.g. policy) can result in them being grouped together when, if the goals are worded differently, they are not.

So what other approaches can be used? In this case we decided to undertake some attitudinal research with users to understand more about how users thought about their goals they had for this site. What this showed was that users initially looked for subject based links (e.g. in this case specific words covered by the categories: health, benefits, death etc) rather than the organisational focused categories being used by the site (e.g. information, products & services etc.).

The original card sorting exercise had not identified this because the goals had not been written in a way that would have surfaced this type of ‘topic’ categorisation. Our conclusion is not that card sorting is ineffective, but rather that you must be very careful about how you write the goals so not to produce spurious associations or exclude other more obvious categorisations.