Karyn: I took a rather circuitous route that started with a passion for psychology. Fifteen years ago, I would have said that I wanted to help people make sense of their life experience. Now, I'd say I want to help people make sense of their Web experience. This interest in psychology barely allowed me to put food on the table so I took a detour and focused on developing practical business skills and experience. Managing a product line of modems got me online with CompuServe before the Web exploded.

My information architecture career started in 1995 when I joined an Internet start-up and began working closely with visual designers and code developers in site development. I moved to IBM in 1996 and got increasingly interested in creating usable, useful and desirable sites. One thing led to another; an Amazon.com addiction resulted in my building quite a library of Web Design, Information Architecture, Digital Culture and e-business books to guide my work.

And yes, I felt insane when I took my current position as an information architect. I have had to approach information architecture in completely new ways just to get my head around the subsite that I'm focused on, www.ibm.com/software.

Lou: With all your business experience, do you ever find yourself returning to your psychology background?

Karyn: Psychology taught me how people think and what makes them feel satisfied and not so satisfied. My MBA taught me about business as a system. When I mix my educational experiences together, my perspective is that information architecture is about people having an experience within a system. This system may provide the best user experience by being transparent to the user-they enter it, do what they need to do-and they may only interact with a small piece of it. But a site needs to be a system (and have a consistent structure) to be effectively managed by a business. This view of how information architecture relates to business is critical in helping management, along with my peers, understand the value of my role.

Lou: So how exactly does one educate one's management and peers at a place like IBM?

Karyn: It takes perseverance and the ability to relate the Web to business. At IBM, the value of having a web site and putting up pages is well understood. But not everyone can see all the holes and problems with an architecture that evolved without planning. I know it needs attention, but if I explain the problems only in information architecture terms, I won't be understood. In this situation you need to understand the challenges that your business faces.

At IBM Software, we are tasked with transforming a huge division into an e-business. So I've taken the time to relate information architecture and e-business; this guides my communication and my activities. My organization has an e-business transformation strategy that guides us all. I studied the Web and e-business projects and their functionality targets to determine which ones need information architecture. A high percentage need information architecture at some level, but I focus on projects where the value is most apparent. Content management is an example of one of our projects that's dependent on strong information architecture because an organization just cannot be effective at managing unorganized content. Showing the "bottom-line" value of information architecture is always a challenge but this approach makes it easier for me to demonstrate cost savings and customer value.

Lou: I haven't heard an MBA define information architecture yet; do you want to venture a definition?

Karyn: As I mentioned before, my understanding (and therefore my definition) of information architecture had to change dramatically when I took on the responsibility for the information architecture for IBM's Software site. I used to think of information architecture as the overall structure of a site and page flows. I usually saw a very limited set of customer tasks and paid minimal attention to the structure of content on pages. But to be effective in my current role, I now see information architecture in terms of two areas:

1. Site structure and organization: Dealing with a huge information space that supports a diverse set of user tasks, it became critical for me to think in terms of providing users multiple ways to access content (by audience, task, geography, product name, brand name, etc.) and to apply these navigation schemes consistently across the site.
2. Page types and content structure: With sophisticated, powerful software products and a purchase process that requires complex decision making, you need to structure pages and page content in ways that support user needs (and reduce or eliminate information overload).

Lou: What's your biggest frustration as an information architect?

Karyn: I don't mind the regular speed bumps that come with being an information architect and working with the Web, but frustration strikes in spades when I see the users get forgotten during the site development process. While IBM has made great strides with our User-Centered Design guidelines and our reliance on usability testing, it doesn't sound like this is the case in many environments. If a site doesn't support user needs, users go elsewhere and companies lose all sorts of valuable things-brand trust, revenues, market share. Speaking in psychology terms, if a site causes cognitive dissonance people remove themselves from the offending situation and don't tend to return.

Lou: What's your theory as to why this even needs explaining in the first place?

Karyn: The Internet is about shifting customer interaction from something that occurs between humans to something that primarily happens between a human (e.g., a customer) and a web site. In such an interaction, what customers need hasn't changed much; how it gets delivered has. Information architecture addresses the how.

Most companies know what their customers want; otherwise they wouldn't be in business. But traditionally companies employed humans (e.g., salespeople) to deliver on the customers' needs. And these humans essentially had an information architecture in their heads, and existed to sort it out for customers.

But in network-enabled businesses you remove the human being from the delivery. As an industry, we're just beginning to understand the impact of this. After all, salespeople didn't just walk into a customer's office, drop a brochure with a list of product features on the desk and then leave. As an industry, our sites need information architectures that meet the needs of users the way a human did in business past. This is most definitely a paradigm shift and something information architects can help address.

Lou: What advice do you have for those entertaining information architecture as a focus? How about for those who find themselves suddenly stuck with the title "Information Architect"?

Karyn: If you are interested in information architecture, make information architecture something that works for you. When I went to the "Defining Information Architecture" conference in Boston I was looking for a definition as if there was only one. I heard so many different definitions that worked for so many different careers and perspectives. For example, when I heard a presenter discount the importance of categorization, I realized that this person just didn't need to address this component of information architecture because his site was small and serendipity was more valuable to his site's users than hierarchy. So I decided to take what made the most sense in my current role and leave the rest until I might use it later. I suggest others do the same by learning all they can and using what they need. One place to start is by reading presentations and materials from the "Defining Information Architecture" conference. Then subscribe to mailing lists that discuss information architecture and Web issues and talk with other information architects to share experiences.

Lou: What sites and resources support you in your work as an information architect?

Karyn: I get a lot of support from drawing on my personal library. As far as Web resources, there are several sites that I visit on a regular basis to keep current and to keep my ideas fresh, I participate in the CHI-Web and SIG-IA mailing lists.

Unfortunately, I have challenges that aren't addressed in these forums, probably because the IBM Software site is just so big. So I'll often use the Web to compare components of my site's information architecture with a couple competitors sites; invariably this teaches me a great deal.

I also find it refreshing to look at the structures of other sites to clarify concepts and techniques. Etoys is my current favorite. After a long day dealing with the complexity of big business software, toys are just such a hoot. But really, Etoys understands the different ways a customer needs to shop for toys and they reflect this so well in their organization and navigation.

Lou: We usually end these ACIA interviews with something fun, and you've already supplied a likely stopping point by mentioning Etoys. But I'd like to end this one differently: what part of your work are you really proud of?

Karyn: That's easy! I'm psyched about a content management project where I've been the information architect: we're using it to make a great ROI case for the value of information architecture. We determined that where we applied content management to just visual design elements, we got an 80% cost reduction during a recent redesign. Where we applied content management to all content elements (meaning everything is published out of a database) we got a 97% cost reduction.

Of course, one of the critical components of a content management system is a strong, user-focused information architecture. Why? Users need content that meets their needs and once you know the details of this, you build a content management system so Web personnel know what to create, where it will placed on the site, what format it needs to be in, when to update it, when to delete it. This was a great project and I'll be thrilled when it leads to my having a bunch of partner Information Architects throughout IBM Software!