Rare University – Session 3, Information Architecture Sitemaps


Instructor:  Adam Polansky




Site maps have been described as arguably the “workhorse” of Information Architecture.  Whether that is true or not, it is certainly the most visible and characteristic deliverable of IA.


What is a Site Map?

Site maps are the logical extension of the need for a visual reference to the relationships between individual pages, content and content spaces.  They are a two dimensional view of the topography of a web site or application. They are driven by content.  While portions of a site map might mirror task flow in a static sense, they are NOT workflow or task flow documents.


Site maps are most commonly used to specify information within new or redeveloped sites but they can also serve to show existing sites or applications where the client is not exactly certain of the current state of a site.  This is common with older sites where many hands have touched them over time.


Site maps as they are created at Rare Medium, provide a great deal of information about the overall structure of a site as well as some indication of the metadata (data about data) associated with specific content.  A casual observer can answer the following questions using a site map:


Small Multiples and 10-5-2 Feet

From a distance an observer can see something about the depth and breadth and pattern of information of a site.  Moving in a little closer they can see the broader content areas and learn something about the nature of each.  Upon closer inspection, they will know any number of things about a particular page.  If a map can provide information at these different levels, it will establish the context that makes all of the information meaningful at any distance.

This approach is an example of the principle of “small multiples”:


“Small multiples work as efficient and convincing summaries of data or an argument, making the same point again and again by offering complementary variations on the major substantive theme.”


Edward Tufte, Envisioning Information.


Reading a site map or any other kind of map is a matter of context. This is what Tufte refers to when he says “substantive theme”.  That context is established at the highest level so that as you begin to focus on specific sections, you understand where you are in relation to the rest of the site as well as understanding why that information is relevant in its location.  To do this we use a set of visual standards and conventions that describe the attributes or “metadata” of a particular page or content type.  In the following course we will use this “Zero-in” approach as a guideline to ensure the integrity of the final design of a site map.


Gathering the Information

For the purposes of this course we are assuming that the information that will populate the map – content types, features, functions, etc. – has all been collected, qualified and quantified.  We will discuss the exercise of gathering and using this list further at the Staff level.  There will usually be a spreadsheet that lists all of what will be the raw material for the site map.


Tools for Site Map Development

There are several tools that work well for creating site maps.  The choice usually lies in the IA’s comfort level with an existing tool.  Regardless of the tool, the following considerations need to be kept in mind:



Examples of the most common tools are:



This course will concentrate on Visio 2000 as the primary tool for site map development due to its easy learning curve, ready-made stencils with complex properties, ease of editing / updating and common industry acceptance.


10-foot view

The first objective in building the map is to establish the context for the information you will be organizing.  That includes the primary and footer/resident navigation scheme.  These elements might be pre-determined by the client or they may be left for you to identify.  They may also be the result of a collaborative effort. 


The site map template is designed to display information in much the same way that the user will encounter it if they follow the primary information paths.  Across the top of the map list all of the primary navigational elements. These are the big 5-7 buttons that represent the main content areas that are the backbone of the site.  They don’t change often and should provide the best, most intuitive inroads to the core content of the site.  When possible place them from left-to-right in order according to some priority this might be in accordance with the client’s wishes or based upon known user goals.  Politics can also influence this display.


The footer or resident navigation is on the top right end of the map.  This is the content that, while not primary to the use of the site, are important enough that they always need to be present.  Examples are “legal notices”, “site map” or links to a parent site.


5-foot view

Once the navigational elements are established, you have a contextual framework against which you can build out the map.  This is either the tedious part or the fun part depending upon your point of view.  The idea is to organize each page or content type within the navigation.  Content or pages might fall easily into more than one content area.  It can become the IA’s determination on where a user would most likely seek that information*.  The IA might still choose to link a page to more than one area or even duplicate it on the map if that seems more prudent.


2-foot view

The last aspect of the site map involves the metadata associated with each page or content type.  Examples of metadata in the context of a site map are:



These attributes and content types are displayed using a set of conventions developed over time at Rare Medium.


(See Stencil Files)


Once the detailed portion of the site map is complete, and reviews and updates have been made, the map is ready to stand as the model for overall development.  Under some circumstances its use can be extended to include assisting content management in requisitioning and accounting for raw content.


For the purposes of Exploration, the end product is referred to as a “site model”.  At the end of prototyping, it will be updated and the final product is then referred to as the “site map”.


Next Session

The next step in the development process is the development of the screen decks and wireframes.





This is a shallow list.  If you check them out you’ll find that they have much more detailed and comprehensive lists of their own.



Information Architecture for the World Wide Web

By Louis Rosenfeld & Peter Morville


Practical Information Architecture

By Eric L. Reiss


Envisioning Information

By Edward Tufte


Web Sites


Argus is the consulting company founded by Louis Rosenfeld and Peter Morville.  ACIA stands for “Argus Center for Information Architecture”.  It is a great source of editorial comment from other IAs, book reviews, interviews and resource information about tools and techniques.



Jesse James Garrett is an IA with Metrius, the web-consulting arm of KPMG.  He has some interesting thoughts on IA that he shares in his offering called “A Visual Vocabulary” He has posted tools and charts to explain his approach.



-Christina Wodke’s wonderful web log (Blog).  The site itself contains reading lists, articles and recommended tools.  She also has a daily email called gleanings that you can subscribe to.  It contains collected articles and information on IA, Usability, Design, and the Web in general and anything else that she feels like sharing.  It’s a great way to keep a flow of information coming your way so that you can always have a peek at what is topical in the IA world without having to hunt it down.