Pat’s Point of View

I am always quite surprised, after one of my workshops, when someone remarks "I didn’t know there was so much to it! I thought IA was just about coming up with the nav".

Whilst I’ve written recently about the many faces of IA, I think there is more to such comments than just a different definition of IA being expressed. Probing a bit further, I found that these remarks were based on a rather superficial view of the work involved. It wasn’t obvious, until I ranted for a whole day, that it takes a bit of effort to whip up the navigation for a site.

There is more to it…honest. Firstly, navigation implies some kind of organisation of the site. But where do we get the labels for the navigation, how do we know what structure is best for the content and what type of navigation will work best?

IA is a UCD practice, that’s user-centred design in that you take input from the actual end-users of the website (or intranet or software) and use that in the design process. I suspect the reason why many people are surprised to hear that IA involves so much work is that they have been thinking of designing navigation as a isolated activity; an ‘expert’ creates it in a matter of minutes and you go off and build it.

That approach is doomed to fail and often results in further costly redesigns, because the end product simply doesn’t work for users. So there is a need to get input from users, both directly using research methods such as interviews, observation and focus groups, but also indirectly through web analytics for example. This builds up a picture of who your audience is, their needs, behaviour and attitudes. You can get their input into labels and content structures, using techniques such as card sorting. This gives you a good idea on how users think of the information they might be presented with, in terms of the names they use to refer to it, but also what they believe should be grouped together. And it can take quite a bit of time and effort to do this properly, especially making sense of what you find.

Then comes the tricky bit, taking what you have learnt through research to design a solution. It could be the overall IA (the site structure), the navigation or the layout of pages. Once you make any design decisions, you also need to validate or test those decisions. To do this you return to the users and seek their direct input again, using techniques such as paper prototyping, Card Based Classification Evaluation and usability testing.

So, yes, navigation is often a key output of IA, but it is just the tip of the iceberg. IA starts much earlier and goes much deeper than this. Attempting to design the navigation—or almost any aspect of a site—without going through the whole process is like working in a vacuum.

It wasn’t my intention to describe the entire UCD process, but I’ve tried to think why people outside the IA field don’t understand that there is more to it. Is it because we practitioners shroud ourselves and our methods in jargon and elitist secrecy? Or is it that so many bad practices are out there, conducted by inexperienced practitioners, that it has tainted the opinion of certain people? Or something else.

It’s not all dire of course, I meet lots of people who are not IAs and don’t have a comprehensive understanding of all the techniques involved, but they do appreciate that to achieve good results you need to invest time in things like research and testing. Yet I’m still intrigued by what people think IA is all about, and possibly more importantly, why they are turning up to my workshop if they think it’s dead easy? :)

(The significance of the picture? Well, these disentanglement puzzles, or ‘blacksmith puzzles’, are often harder than you think. Just like IA might be to the uninitiated. And it was less of a cliché than a ‘tip of the iceberg’ image.)

Popularity: 59% [?]