Peter Morville and Lou Rosenfeld
Published on December 10, 2002
Digital Web: We hear about Little IA and Big IA; tell us about these definitions and their differences.
Lou Rosenfeld: "Little IA" means doing the stuff no one else wants to do, such as inventorying and analyzing zillions of pages, identifying content objects, and developing controlled vocabularies to describe them.
"Big IA" is the stuff that everyone else wants to do, including interaction design, project management, visual design. These are the fuzzy areas which lead to the counterproductive posturing and bickering that can really drag a team down.
We can minimize these arguments by remembering that the lines are more likely to be drawn based on practical considerations--such as who's on the team, what skills they actually have, how much time they have, and what the project entails--rather than some sort of textbook definition of what information architects, interaction designers, and others actually do.
If readers want to delve a little deeper, Peter did a nice short piece on this distinction back in July of 2000.
Digital Web: We also hear that IA is design and then we hear IA is not design. Which is it? Why?
Peter Morville: We've always recognized that information architecture involves a balance of art and science. We can study users and usage. We can survey stakeholders. We can analyze content. But, in the end, we must listen to gut instinct, be creative, and take risks. I can't imagine anyone making a convincing argument that this conscious act of shaping an information space is not design.
Digital Web: Where does IA play a role in the Web development process?
LR: The answer really depends on what's being developed. If you're starting a site from scratch, or facing a complete renovation, then information architecture is a major component of the plan that should guide actual development. In these situations, you need to determine right away who the site's main audiences will be, what kinds of information needs they'll have, and how they'll navigate the site.
You'll also want to identify the content you have and the content you'll need, and how you might make that content accessible to users. Finally, you'll want to ensure that the site's mission supports the organization's mission, and that your designs can actually be implemented given the organization's resource constraints and political barriers.
This is a very top-down process, and is typical of the kind of information architecture that was practiced a few years ago, when we were more likely to be designing brand new sites. In such situations, information architecture provides some of the methods used in an initial research phase, and yields important deliverables--site maps, wire frames--during the design phase.
Nowadays, we're more likely to be saddled with huge and disorganized information environments that are impossible to navigate once you get beyond the upper layers of the site.
So we find ourselves working on the depths of a site--structuring, classifying, and linking pockets of content that may be owned by different business units and serve different audiences. This type of bottom-up work is more like fixing the plane while flying it: you don't have the luxury of the three standard project phases--research, design, and implementation--to fall back on. This kind of work happens when it can: when you, as the information architect, has made a successful case for it, or when the need is so painfully obvious that the organization has already bought in.
Digital Web: What are the typical tasks involved in doing IA?
PM: Long Answer: An information architect must learn about business goals and context, content and services, and user needs and behavior; and then work with colleagues to transform this balanced understanding of the information ecology into the design of organization, labeling, and navigation systems that provide a solid but flexible foundation for the user experience.
Short Answer: Read our book. :-)
Digital Web: People often get IA and usability confused. Can you clarify the differences and explain how IA relates to the various aspects of Web design, usability, and navigation?
LR: Another short answer: IA is about findability, and usability is, well, about usability. An information system can be usable, but that doesn't guarantee that it'll help you find information you need. And vice versa. I highly recommend Jeff Lash's recent column on the difference, published right here in Digital Web Magazine.
This confusion is just another illustration of how the recent IT revolution has completely muddied the world of information system design. Many usability engineers who started their careers with a focus on system testing and evaluation now find themselves drawn to the joy of design. Conversely, many information architects and others who focused on design are now looking to fields like usability engineering for methods to validate their designs.
The market for our services is similarly confused. For example, we often find that prospects who ask for usability services are often really seeking design solutions, many of which involve information architecture.
Whether we realize it or not, we're all searching for a shared vocabulary with which to articulate the problems we're having designing, maintaining, and using information systems. That's why any design-related professional needs to be ready to educate clients and colleagues alike. Providing that shared vocabulary is half the battle.
Digital Web: What is the purpose of doing a content inventory? What's the typical process of doing one?
PM: Building an understanding of the nature, scope, and volume of existing content is an important step in the redesign of any information system. It's a vital component of the bottom-up approach to IA design and an area we cover extensively in the second edition. Early in the process, content analysis can provide insights regarding ways to structure, organize, and provide access to information and services. Later in the process, a content inventory or audit enables construction of a detailed implementation plan.
Digital Web: Can you give us an example of what a site with good IA would look like? A site with bad IA?
LR: I used to think that IA was noticeable only when it was bad. Amazon proves that assumption wrong. Its information architecture deftly balances the needs of users who know what they're looking for with Amazon's desire to make recommendations for further purchases. Amazon has carefully selected different means of connecting users with appropriate content, ranging from comprehensive searching and browsing to selective recommendations based on collaborative filtering and user-supplied reviews. You could argue that IA is the main ingredient in Amazon's brand.
At the other end of the spectrum, I love to take shots at Brint.com, a site with too much information architecture. It's as if the site's designers got so excited about the almost infinite number of ways they could connect users with content that they decided to use them all, regardless of the overhead associated with maintaining them. One of the main goals of IA is to pick the few best architectural approaches that will serve the most (or most important) users without bankrupting the sponsoring organization.
Digital Web: What are the obstacles to constructing a Web site with a solid architecture? How do you overcome them?
PM: The biggest practical obstacle is that most Web sites are built by people who have little understanding of the basic principles of information architecture or user experience design. They don't know what they're doing and they don't perform user research or user testing.
Of course, that doesn't mean it's easy to design a useful, usable Web site, even when you've got an experienced team on the job. Major obstacles include:
- When the incentives of powerful stakeholders within the company aren't aligned with the goals of shareholders, employees, and customers, it's impossible to design a successful, sustainable solution.
- Even under the best of circumstances, the inherent ambiguity of language and organization make it extremely difficult to design shared information spaces that support usability and findability.
- A solid architecture may not be an adaptive architecture. In a fast, changing world, we can't simply optimize for the present. We must design agile, scalable architectures. And that's not easy.
Digital Web: What do you see as the main issue(s) with intranet design? What key things should be considered when designing an intranet to ensure its success?
LR: Intranets tend to grow organically, with silos of content that correspond to the corporate org chart. Unfortunately, users usually don't have that org chart in mind when they're looking for information to help them do their jobs. They are thinking in terms of topics, tasks, and processes, or identifying themselves with specific audiences or roles. But most intranets require users to think first about which business units and departments might provide the answers, and then look in those places.
Well-designed information architecture, combined with strong content management practices, can enable users to cut across the grain of these vertical silos of information. So when an employee is seeking information on booking a work-related trip, she can find it one place, instead of hunting through the HR site for policy information, the IT site for a travel-booking application, her own department's site for an approval form, and so on.
Of course, this is all easier said than done. There are huge political and cultural issues to deal with inside any large organization, not to mention extensive political and cultural issues to contend with. And management often expects these kinds of problems to vanish in a ridiculously brief time period with little thought about organizational implications.
While we don't have all the answers, our new edition does provide a framework for figuring out how to address these sorts of "enterprise information architecture" challenges in your own organization. We also provide a case study to show how Microsoft developed both a new information architecture and a new technical architecture to address these problems for its intranet environment.
Digital Web: How can we measure the effectiveness of IA in terms of ROI?
PM: In our second edition, we have dedicated an entire section to "Making the Case for Information Architecture." We explore a number of ROI arguments, examining:
- the cost of finding information
- the cost of not finding information
- the cost of frustrating users
But we also explain why ROI is not enough and suggest other ways to measure effectiveness and show value. For example, we delve deep into the realm of business strategy, exploring how information architects can contribute to strategy formation and the development of sustainable competitive advantage.
Digital Web: What are the clues that demonstrate a site has obviously put a lot of thought into its IA?
LR: I believe there are five critical junctures where the large majority of users' experiences with a site's information architecture will take place. I look at these areas to see if much thought has been invested in the site's architecture:
- Main page: Does it anticipate and answer the major questions that users have when they arrive at the site for the first time, does it link to a site map to provide an orientational overview of the site's content? Does it do the same thing for repeat visitors?
- Primary browsing interface: Does the site's navigation system provide both broad and deep access to the site's content? Do the labels match the language of both users and content?
- Search interface: Is it easy to find and consistently placed? Do advanced functions come up at the right moments, or are users always presented with too many complicated options?
- Search results: For each result, are users provided with the right kinds of content to enable them to assess its relevance, or do they always have to click through and look at the actual document to make a decision? Are search results ranked or clustered in a way that matches users' needs?
- A document's contextual navigation: Now that the user has demonstrated a specific interest by navigating to a particular document, does he receive specific, appropriate advice on where he can go next?
If you spend a little time on these five areas, you'll glean significant information on how well designed the architecture is and what can be done to improve it.
Digital Web: Many have stated they believe IA is only for content-rich sites like magazines, news, and search portals. Do you get the same impression from people on this view? What are your thoughts? Why should IA be done on other sites?
PM: Our personal backgrounds in library science and our passion for findability have led us to focus a good deal of our energy on content-rich sites. However, many other information architects focus on the design of highly interactive Web sites and software products where navigation and task completion go hand-in-hand.
LR: And let's not forget that even in an application-rich site, you still have to find the right applications before using them.
Organizations, Businesses, and Uses
Digital Web: AIfIA (Asilomar Institute for Information Architecture), of which you're both board members, has just launched. Why was there a need to develop this nonprofit organization?
LR: As a community, information architects have never had a home. Although ASIS&T has been very supportive of the field, no single organization was there to ask the community what it actually needs, much less to promote IA outside our community, assemble materials to improve IA education, produce community events, and support local IA activities.
And similar community-building efforts, whether sponsored by for-profit companies (Argus' ACIA) or that were completely volunteer-driven without any membership structure (info-arch.org), haven't been successful. So we're trying a more traditional association model, and so far, so good. After a few weeks, we already have about two hundred members. We hope that readers will at least visit the site--http://www.aifia.org--and consider joining and participating in one of our initiatives.
Digital Web: What are the goals for the organization? What will AIfIA do that is different than what has already been done by others?
PM: AIfIA is a non-profit organization dedicated to advancing the practice of information architecture. To our knowledge, AIfIA is the first and only international professional association created specifically to address the needs of the emerging information architecture community.
Digital Web: AIfIA was largely promoted to the Web design crowd, a group that is already familiar with you and your book, Christina Wodtke's efforts, and the importance of IA. Why promote to this crowd, which is like preaching to the choir? What about targeting businesses, management, and people who are involved in Web design, but not necessarily up to speed on the important factors like IA and Web standards?
LR: Practitioners are only one of the audiences we plan to address. The others are business decision-makers, other new media people, and the press. We're taking a grassroots approach to promotions, with the goal of putting promo materials into the hands of practitioners so they can spread the word themselves. Jeff Lash is heading our promo team, and anyone interested in helping out with promoting the field of IA should contact him.
Digital Web: Congratulations on releasing the second edition of the Polar Bear book. What were some of the biggest surprises or discoveries in writing the second edition?
PM: Thanks! The biggest surprise was that we had so much more to say. The second edition is almost 500 pages, more than twice as long as the first. In many respects, it's a completely different book, though you can't tell that from the cover.
Digital Web: Now for the whiz-bang finish: Why do you feel IA is not being used where it should? What evidence is there to support this? How come few take IA seriously? What makes you believe that people who don't take it seriously should bother with IA at all?
LR: Actually, I'm hugely optimistic about the field. I don't worry about IA not getting taken seriously. A few years ago, no one had heard of IA, yet nowadays I keep hearing about other design professionals who are hysterical about the threat they perceive from information architects. Moving from blank stares to red faces implies that IA has arrived, although we still have a lot of explaining to do.
I don't get frustrated about where IA isn't being applied, because I know that the situation is far better than it was five or ten years ago. In fact, there are lots more information architects than ever before and most of them work in-house. I can personally guarantee that it's far easier to sell IA today than it was in 1995.
Though things are tough now, time is on our side. The information explosion isn't going to stop or even slow down. The good content we have today will be stale tomorrow, and technology won't solve all of our problems. So information architects need to take a deep breath, be patient, and look at the long term. There's absolutely no reason not to be optimistic.
Peter Morville is President and Founder of Semantic Studios, an information architecture and strategy consultancy. Peter is also the co-author (with Louis Rosenfeld) of the best-selling book Information Architecture for the World Wide Web. A second edition of the book is expected to be on bookstore shelves in August 2002.
Lou Rosenfeld is a co-author of Information Architecture for the World Wide Web (AKA "the Polar Bear Book") and principal of Louis Rosenfeld LLC. He has been instrumental in helping establish the field of information architecture, and in articulating the role and value of librarianship within the field.
Meryl K. Evans, content maven, is a WaSP member even though she's far from being a WASP. The content maven writes a column for PC Today and blogs for the Web Design Reference Guide at InformIT. Meryl provides the home for the CSS Collection and she's the editor of Professional Services Journal, meryl's notes :: the newsletter as well as other newsletters, so tell all your friends, families and animals to subscribe. Her ancient blog keeps cluckin' since its arrival on the web in 2000.