How can we make ever-growing volumes of information accessible and useful to people without overwhelming them?
That is the question I want to consider in this third and final instalment on information wayfinding. In Part 1, I argued that we must move beyond thinking of information interaction as a book of pages, and instead think in terms of a spatial environment. In Part 2, we compared information interaction to the process of finding one’s way through a city, and then defined three elements of the information environment: districts, layers, and nodes.
Building upon this foundation, I would first like to consider how people move through information environments. We’ll then be equipped to outline a new set of guidelines for building websites and applications that enable people to make sense of ever-growing volumes of information without overwhelming them. In particular, we’ll look at integrating search and browse into a single, seamless experience.
1. Wayfinding Behaviour
Information wayfinding provides a valuable lens for viewing humankind’s interaction with information. The term wayfinding itself refers to the collection of cognitive processes people use to navigate physical environments, what can be summarised as spatial problem solving . Information wayfinding, then, is the collection of cognitive processes people use to navigate information environments.
“Information wayfinding, then, is the collection of cognitive processes people use to navigate information environments.”
Information wayfinding is most often exhibited in one of three different modes: locate, explore, and meander. Many readers will immediately recognise the connection with Gary Marchionini  and David Elsweiler’s  work, which is also summarised in my and Tony Russell-Rose’s book Designing the Search Experience.
People operate in the locate mode of wayfinding when they know precisely what it is they’re looking for, but need help finding where to look. These are sometimes called “lookup” queries, and there is typically one right answer to the question. A few examples of locate needs might be:
- “Who was the director of the film ‘The Third Man’?”
- “What is the population of London?”
- “When are expense filings due?”
- “Where is Jimmy’s Pizza”
Exploring is much more open-ended than is locating. When people explore, the journey is as important as the destination. As people encounter new information, their information need evolves, resulting in an iterative, ongoing process.
Marcia Bates has described this process as “berrypicking” — people move from one source to another picking up nuggets of information along the way . Peter Pirolli and Stuart Card describe the process as “information foraging,” where people follow “information scent” as they go . Both of these models depict information seeking in spatial terms, and reinforce the concept of information wayfinding as a result.
Exloratory wayfinding might be initiated by questions such as:
- “What kind of car should I buy?”
- “Where should I go on holiday?”
- “What film should I watch this evening?”
While locating and exploring are both prompted by an information need, there is a third category which is not: meandering. People aren’t always looking for something in particular; they may also have motives such as having fun or killing time . Checking email or social networks while waiting for the bus to arrive is just one example of meandering behavior.
These three modes — locate, explore, and meander — are not new ideas, but they come to life and gain new meaning when placed in the context of information wayfinding.
2. Guidelines for Optimising Wayfinding
Once we start thinking of information as an environment and wayfinding as the means by which people interact with that environment, the obvious next question is: how can I apply these ideas to the information environments that I play a part in building?
“How can I apply these ideas to the information environments that I play a part in building?”
There are six key principles that should guide how we construct information environments: structured districts, flexible layers, positional cues, survey knowledge, clear paths, and coherent interaction.
To make each of these principles more tangible, I have created an example using Twigkit, my company’s software for rapidly building search-based applications, and the Google Search Appliance. It uses a collection of films to demonstrate what an information wayfinding experience could look like. (I also recorded a 5-minute video about how the application was built.)
2.1 Structured Districts
We’ve said that districts are the main categories into which an environment can be divided. Logical, clear districts are important to help users understand the information environment at its most basic level. Districts should correspond with the user’s own mental model. When people think about films, for instance, they think in terms in genres: a film is an action film or a drama film, a comedy or a thriller. Virtually every domain has some dominate organisation scheme that should form the districts of the information environment.
Districts should be clearly presented to users, perhaps in a form resembling traditional web navigation. Correspondingly, the URL scheme should also be based on districts.
2.2 Flexible Layers
In addition to districts — which serve as the primary organisation scheme — users should also be able to filter by secondary criteria, or layers of the information environment. The flexibility of layers is a vital companion to the rigid hierarchy of districts, and the two should operate in tandem.
2.3 Positional Cues
Like a “you are here” marker on a map, positional cues help users orient themselves within the information environment. More frequently, positional cues take the form of breadcrumbs — they indicate the district in which the user is browsing, what layers they have applied, and any search terms they have entered. These positional cues give users a sense of security, and avert any feelings of lostness.
2.4 Survey Knowledge
Helping users gain survey knowledge is a second form of enhancing orientation. But while positional cues more often depict where the user has been, survey knowledge helps them ascertain where they should go next. Like studying a map or looking at the landscape from the top of a mountain, survey knowledge gives users an overview of the information environment. Data visualisation — from subtle indicators to full fledged charts and graphs — are an excellent tool for enhancing survey knowledge.
2.5 Clear Paths
In a physical environment, having a clear line of sight to the destination greatly simplifies the act of getting there. So in information environments, having clear paths ensures that users always have a path onward.
A good example of this principle is having links to related content. When viewing an information node (what we often call a “detail page”) for a film, for instance, having clear paths to similar films, to films by the same director, starring the same actors, and so forth, gives users a sufficient number of paths forward.
2.5 Coherent Interaction
Most importantly of all, I believe that interacting with districts, layers, and keyword search should be a single, seamless experience. Layers should not only be available in “search mode,” and keyword search should not be divorced from districts. Instead, users should be able to navigate to districts, apply layers, and perform searches iteratively and conjunctively.
To make ever-growing volumes of information accessible and useful to people without overwhelming them, we’ve seen that we must move beyond the page metaphor, think of information as an environment, consider how people interact with that environment, and finally, apply design principles for optimising wayfinding.
In fact, if we were to survey the world’s most successful online properties — Google, Facebook, Amazon, eBay — we would see that they have already embraced the principles of information wayfinding.
The question is: will you?
 Arthur, P. & Passini, R. Wayfinding: people, signs, and architecture. McGraw-Hill Book Co., 1992.
 Marchionini, G. (2006). Exploratory search: from finding to understanding. Communications of the ACM, 49(4).
 Elsweiler, D., Wilson M. L. and Kirkegaard-Lunn, B. (2011) Understanding casual-leisure information behaviour. In Spink, A. and Heinstrom, J. (Eds) New Directions in Information Behaviour. Emerald Group Publishing Limited, pp 211-241.
 Bates, M.J (1989). The design of browsing and berrypicking techniques for the on-line search interface. Online Review, 13(5), 407–431.
 Pirolli, P. (2007). Information Foraging Theory: Adaptive Interaction with Information. Oxford University Press.