News 

Can you start off with the elevator pitch for Information Architecture?

I'll give you the boring definition first. It's 'the Art and Science of structuring, organising and naming information so that it's easier to find and to manage.' We are talking about information structures - not just their shape but also the symantic value and where it comes up in the hierarchy of your site. Information architects have to balance the needs of two different audiences, users and the people managing that information.

Information Architecture is about finding and we are 'findability engineers'. We need to find what users are trying to do - searching, browsing, even asking questions are all part of finding. Sometimes its good to think of these separately but ultimately information architecture integrates them because that's what people do. They integrate through many iterations of searching and browsing so a site should support that.

Different business models mean different information architectures. Dell and Hewlett-Packard may have pretty well the same users and on their sites you will see basically the same content - technical support, range of products and so forth but the business models are totally different. One is selling through resellers, the other is selling through the site. I divide the world into three circles, content, users and context. There are so many, you are always going to find some aspect of one of those circles being very different.

How do you sell IA to the guy who signs the cheques? In your book, there is almost a running theme about how Information Architecture strategies get caught up in company politics

Yes...That is the biggest change between the two editions. Its one thing to think about content and users and how you connect them but that's two legs of a three legged stool. The third leg is the business context - navigating business, culture, resource constraints and strategy so that your concept can have strategic value. We are trying to position ourselves as people who can make problems go away. So we not only come up with the architecture but also the policies, procedures and strategies. Rather than talk about information Architecture, let's get the problems out on the table and come up with the solutions.

Today, Information Architects are less likely to be dealing with new websites than old ones which have become overgrown with content as different parts of the site do different things. Where do you begin in this situation?

I hear that from a lot of people, this feeling of being overwhelmed. The content is out of control and there's a lot of politics involved because business units are off doing their own thing. What we do is break up this large, intimidating, messy problem into small pieces that can be taken

ADVERTISEMENT

on. So, instead of trying to get everything in order at once, you need to think about the content areas with highest value for users and the highest strategic value to the organisation. What I am talking about is the low hanging fruit that can be used as a model of what can be done over time. Instead of a strategy that might take three to six months to get everything under the roof, you think about pockets which can be organised as part of a three to six year process.

Don't we still need a grand vision to begin with?

You are not talking about how do we think about it architecturally. When most enterprises start to grapple with these problems, they do a very top down architectural effort. Often they will build a portal at the very upper levels of the site and that might fit in for the look, feel and behaviour of the site. But it's skin deep and, when you drill down, you still have the same garbage so you also need a bottom up architecture which looks to exploit what we know about the information on our website, the destinations, what we know about users and how they find their way to their destination. What I'm talking about is contextual navigation. For example there are a range of British e-government sites, but I'm interested in Portsmouth. In order to find that information, I have to go up the top of one site and down and up another site and down. Wouldn't it be better if I could just have another view that just said "Portsmouth"?

So you look to develop a horizontal navigation rather than a vertical one?

Exactly. The grain is currently vertical and we need it to cut across to a more user-centric approach. Now we can start making links between content at the bottom and not require people to hop up and down the hierarchies.

What are the top five things that a person implementing an Information Architecture on their site should be thinking about?

In my opinion there are five critical junctures where users interact with their information architecture, the five places that you should really think about. These are, the home page, the search interface, the search results, how they are presented, whatever the interface is and then the actual documents that are found and what kind of contextual navigation they may have, how they link to each other. Those are the five places to really consider. The mistake that people make is that they get really fixated over the home page and it becomes a real political battle.

What do you see as the future for information architecture?

What makes me comfortable with where Information Architecture is going is content. I often quote the saying that in the last five years, more information has been recorded than in all the rest of human history. With all the new technologies we have, we are dealing with an exponential growth in content. The other problem we have is that the content we have today may be very good. It may be appropriate, accurate and authoritative. Give it six months, its not any good any more and its sets into something we call 'rot' - redundant, outdated and trivial. That's what happens to content when we leave it alone. So we need to be vigilant not only with exponential growth but vigilant with what we have. As that problem becomes clear to everyone, I'm confident that I'm not going to have to sell this stuff as a concept but as a specific type of service.