Information architecture covers all aspects of design, from the most abstract to the most concrete. Everything from thinking about the content of the site, right down to embedding a navigation strategy, determines the structure of a website. Once that structure is in place, it begins to shape and guide the design process. In this podcast, Dan joins Adam Churchill to address the questions he didn’t have time to answer during his virtual seminar.
Adam Churchill: Welcome everyone, to another episode of the SpoolCast. Late last fall, in a special online seminar, Dan Brown offered up five of his principles to improve your information architecture. Dan and the folks at EightShapes are working with us to create more of these seminars throughout 2011.
Information architecture deals with all levels of design. It deals with everything from abstract mental models and concrete navigation labels... This breadth ends up yielding a range of principles, from those dealing with how to think about content to those that suggest how to embed navigation into page designs.
In his 90-minute seminar, EightShapes principal Dan Brown took a tour of five of the principles that he recommends to shape your design work. And he's graciously offered to come back and tackle some of the questions that we didn't get to address in that seminar.
Hey, Dan. Welcome back.
Dan Brown: Thanks, Adam. It's great to be here.
Adam: Now, folks that didn't join us for that particular seminar, you can still get at it. Like all of our seminars, they're in the UIE User Experience Training Library on our site. Right now, there's over 55 recorded seminars from experts just like Dan Brown.
Dan, for those listening who weren't with us for that presentation, can you give us an overview? Can you tell us about the principles that you spoke of?
Dan: Sure, I'd be happy to. As you indicated, these are principles that I've just collected over my expierence, over the last decade and a half, working on a variety of websites. And Information Architecture is one of those things that covers all manners of design, from the most abstract to the most concrete. So these principles really work the range.
As far as the more pragmatic ones, I talked about three principles, principle of choices, front doors, and Exemplars. Choices deals with how many options to present users on different screens, different navigation elements. Front doors deals with the idea of, users are coming at your website through lots of different pathways. And then, Exemplars deals with the idea that it's important for us to use examples of content when we're explaining categories. So those three principles are structural in nature, but they also focus on how we represent structures in our websites.
Two principles that I covered are a little bit more abstract, the principle of growth and the principle of focused navigation. Principle of growth deals with techniques for thinking about how a website is going to expand and evolve over time. And then, the principle of focused navigation deals with a personal pet peeve I have, where navigation systems tend to be a little mixed-up, that people tend not to think in terms of what the purpose of their navigation systems are, but instead, think about it in terms of where they appear on the page.
So the idea behind focused navigation was to empower information architects to think more about what the purpose of each navigation menu is within their overall navigation strategy.
Adam: Excellent. Let's get to some of those questions that were left over. There were a bunch of good ones. So Dana asks a question about how information architects think about their navigation strategy. And how does it interact with content strategy?
Dan: Ah, the hot button issue of the day: how information architecture and content strategy relate to each other. I think it's really important to recognize that these are intertwined disciplines. Especially when thinking of content strategy within the context of a website. Both IA and content strategy can deal with lots of different media.
Erin Kissane just posted a blog post to Brain Traffic's blog about how content strategy really expands beyond the idea of user experience. And I think it's a worthwhile topic to explore. What are all the different facets, maybe not facets, but the different areas in which content strategy can play?
But when we're talking about a website specifically, there's obviously a huge connection here. And I think the balance between information architecture and content strategy, when we're talking about navigation systems specifically, goes beyond just the labeling of categories. This is important.
As an information architect, I like thinking about, conceptually, what are the important categories that belong to a navigation system or a menu. I'm concerned about the labeling, and that's certainly an ingredient of information architecture. But more and more, I am leaning on content strategists, who have their finger on the pulse of the overall tone of the website, to adequately describe the concepts behind the navigation categories.
At the same time, there is analysis that happens within the content-strategy umbrella, for lack of a better definition or categorization. There's analysis that happens there that can inform the navigation structures. Such that, a content strategist is intimately familiar with the website's content, has a deep understanding of how it relates to other kinds of content, and really can provide useful advice on whether navigation categories are attempting to mix apples and oranges in a way that would not be meaningful to users.
Adam: Kevin wants to know when search-based navigation should be leveraged. And what do they need to be thinking about for it to be effective?
Dan: I'm guessing that he's referring to navigation that appears after someone does a search. The example that comes to mind is on Amazon.com, if I conduct a search there are facets that appear that allow me to narrow down my search results. I think from the information architect's perspective, the important concept here is, that navigation menu, that navigation system is focused. Right? It has a purpose which is to narrow down search results.
And the things that we draw into that... That is to say, how we figure out what should go there, is really driven by the underlying meta-data. By meta-data, I mean the system that we have for classifying information.
The role the information architect plays is trying to understand what's going to be the most meaningful way for users to narrow down that content. One example might be that, if users conduct a search on author name, that using author name after narrowing down is really not going to be meaningful because, presumably, there's only one author in the search results.
But if the author has done both works of fiction and nonfiction, or if the author has written books by himself versus writing books with others, written books on multiple topics... All of those are nice ways to aid the user in whittling down those search results without forcing the user to do multiple searches to narrow it down manually, so to speak.
Therefore, the IA needs to play a role in what that underlying meta-data is. And I think an information architect would think differently about meta-data if he or she knew it was being used as part of a dynamic navigation mechanism for whittling down search results, as opposed to meta-data that's merely used, say, for the management of content.
So thinking through the labeling and controlled vocabulary of the meta-data that's going to be bubbled up into search-related navigation, it's a different approach. We can't be as lenient with those things, because they will be exposed to users of the system.
Adam: The group at Lokian was looking for some examples, or possibly even principles, maybe ones that you spoke of in your seminar or maybe ones that you didn't, around real-time user-driven navigation systems. And they're looking for examples that reveal categories based on current activity, either in the individual or collective.
Dan: It's hard for me to think of effective examples, I mean, tag clouds were really big for a while. And I think we saw almost an immediate backlash against them. And I haven't seen any really good uses of them. I appreciate designers continuing to experiment with them. But I'm not confident that they're offering a lot of value to the target audience.
The other example that they asked about is elevating popular content. And this is definitely a pattern that I've used in my own work. I think we see examples of this on news websites, like CNN or New York Times, where they have some algorithm for determining what is popular content, and based on that algorithm, then show what the top-10 items are, be it the thing that's shared the most or the thing that users click on the most.
The obvious critique is that there's a self-fulfilling prophecy. By virtue of exposing the link as popular, we make it more popular because more people click on it. Because of that, one has to wonder if it's genuine or honest or even ethical to promote these as popular.
At the same time, I think the purpose of escalating this content is simply to have exposed users to the range of content.
I've never necessarily been involved with determining exactly what algorithms are used and I suspect they're becoming more and more specific in how they escalate certain content and hopefully eliminating some of the bias. That being said, I think it would be more interesting to me, not necessarily see content that's popular, but also see content that's curated based on my interests, that is to say, content that the editors think I would be most interested in, or because of other things that I've clicked on.
In a sense, creating a more personalized view rather than aggregating across a whole pool of users who may or may not share my interests. Regardless of whether user generated navigation is used or not, it always should accompany a more defined navigation strategy. That is to say, it shouldn't be the only navigation system in your navigation strategy.
I think it does not yield a convenient browsing experience because it's not necessarily comprehensive. And I think there remains of the desire as users navigate content to see the range and depth that's out there. That range and depth is not always reflected in what's popular.
Adam: Cynthia was hoping you could speak a bit more about the exemplars in an intranet. Any specific examples that you've come across?
Dan: Well just to reiterate, the principle of exemplars says that the best way to describe a category is to show examples from that category. And I don't see this as much online as I would like to, honestly. I think it's very powerful but the tendency these days to streamline and remove content from interfaces has yielded categories without descriptions.
That being said, on a intranet it's especially powerful because... Especially in large organizations where different things can still mean different things to different people, escalating examples from different topics will be worth while.
I mean, imagine you are in one area of the organization and you're interested in learning about another area of the organization. Trying to escalate key policy documents or other files that are germane to that part of the organization can be a great way of learning about it. One of the ways that I prefer to organize intranets is by content type. Because people generally show up having a better sense of the kind of document that they're looking for rather than the topic that it might be found in.
That's not to say one should ignore topics on an intranet. But I really like exposing content type based navigation, policies, forms, reports, things that describe the structure of the document. And obviously, in there, escalating examples is worth while. Now that I think about it, combining that sort of thing with some sort of algorithm to focus on content that is popular or most useful, or in some other way is maybe most recent, can provide a nice description of that category.
So what I'm getting at is, under forms, we know that a frequently accessed form might be a service request that had to deal with a problem with your equipment, or a vacation request form.
So escalating those things because we know that they're frequently accessed is a great way of explaining what's in here, what you're going to find in here, but also providing quick access to the things that people use most.
Adam: So there's a question that wants to know your thoughts about users always being able to answer the question, "Where am I?" What do you have to say about that?
Dan: Back when information architecture was first forming a theory of information architecture. One of the things, I think we all pretty much agreed on is that, any given web page on a website should answer two questions, maybe three questions for users. "Where am I in the whole scheme of things?" "Where can I go from here?" and maybe, "What's the overall range of things that are available to me?"
I think one of the reasons why I appreciated this question is I think the nature of that question is changing, where it becomes less about "where am I," in relation to the overall website which may be too big to really get our arms around and maybe more about "where am I," in the local neighborhood of content that we have. So another way of thinking about this is I've been, in the back of my head thinking about the relationship between navigational structures and an urban environment.
What's interesting to me about urban environments is that they have this fractal organization where, as you zoom in, you see the same patterns over and over and over again, but at a smaller and smaller scale.
And I think we can say the same thing about navigation, especially as websites become much, much bigger that it's difficult to represent the full range of information that's on a website. But at the same time, we can represent the range of information in this neighborhood, so to speak. I think it's important to answer those questions at a neighborhood level. Let me see if I can think of a good example.
Well, let's say on a high tech website like an IBM or an SAP, we've done work with very similar clients where they deal with lots, and lots, and lots of different kinds of high tech products and services and all manner of things that they might sell. And it's really hard to represent the breadth of the things that they sell on their marketing websites on a single web page.
Again, this trend to streamline the amount of content to try and limit the number of choices really gets us to say, when someone lands on a page about a particular product, let's give them a sense of what are the other products nearby and use that product as a filter for the whole range of products and really zero in our neighborhood based on that.
My partner, Nathan, calls this a greater emphasis on local, so focusing more on what's immediately relevant to me because I'm looking at this page. We shouldn't deny the user escape hatches. We shouldn't deny the user, in a sense, an airport or a train station that would take them to a whole other city if they wanted to.
But at the same time, we shouldn't focus on trying to spell out the entire Amtrak system for the northeast corridor but, instead, give users the sense that if they need to get out of DC, they can go to Union Station.
So to be more concrete about that metaphor, I need to convey to users that there are other places to go on this website. But the bulk of my navigation is really going to focus on what's immediately relevant to the content that I'm looking at right now.
Adam: There are couple of questions that came in from RC about the "fat footer." What are the pro's and cons? And in your opinion, what are the key characteristics of successful "fat footers?"
Dan: I have to say I like fat footers except for the name. I think it's an unfortunate name. But I think what's great about a footer that provides a lot of detail is that it calls attention to the information architecture of a website. I've designed a few of these now and I think the value to them is that they can provide that picture of the overall urban landscape.
So I'm in a certain neighborhood, but maybe I want to see what's in the neighborhood next door, or I really want to get a sense of, "How does this whole city work?" The challenge is an editorial one, frankly. It's one in which you don't want to do too much. This may be the opportunity to break the principle of choices where we do want to display a lot of choices, but there's still a balance to be had such that a fat footer still needs to be scannable, so that I really get a sense of the kinds of things I'm going to find in here.
So we generally have headers with navigation categories underneath those headers, so it's not just a laundry list of navigation categories. We still see a tight focus on those areas of the city where we think, the main tourist attractions, if you will, the main areas of the city where we think people are going to want to land or want to go to with out necessarily spelling out the detail of all of those different neighborhoods.
I think, generally speaking, I don't want to render advice in terms of the screen or pixel dimensions because that can really vary by platform. But too much more than a screens-full or even half of a screens-full of a fat footer really ends up defeating the purpose of what it is. Which is to paint the picture of the overall system and provide that escape hatch to other crucial parts of the content ecosystem, if you will.
Adam: Well, that was great, Dan. Thanks for circling back with us and getting the answers to those questions. To those listening in, thanks very much.