Website structure

The information architecture of websites includes an organisation scheme (content categories) and a structure (usually hierarchical).

Organisation schemes

Information can be structured according to a variety of organisation schemes.

• Alphabetical: phone directories, dictionaries
• Chronological: diaries, event calendars, publication archives
• Geographical: travel, news and weather
• Topical: yellow pages, newspapers, academic courses
• Task oriented: menu systems on desktop applications, e.g. file, edit, format menus
• Audience specific: students, staff, visitors
• Metaphor driven: desktop computers use several, e.g. file and folder system, recycle bin

On the web, hybrids of two or more of these schemes are usually used. For example, on the Monash home page a hybrid of audience-specific and topical organisations schemes is used.

Organisation schemes work best when users understand them. Work with target audience groups when creating or refining the site's organisation scheme. Card sorting and content agreement analysis (see resources below) are user-centred methods for designing an organisation scheme. Usability testing can confirm that the resulting scheme is workable.

Avoid designing an organisation scheme that mirrors the structure of the organisation. Users who are unfamiliar with the organisational structure may find it hard to locate resources. Also, organisational changes will require changes to the organisation scheme.

Organisation structures

There are two main approaches to organisation structures on the web: hierarchical or top down, and bottom up as in database models. Some sites also use a linear structure, though this is generally unsuitable for anything other than very small sites.

Hierarchies (top-down approach)

Hierarchies are simple and effective ways of organising information and everyone is familiar with how they work.

Three important things to remember about designing hierarchies are:

• hierarchical categories are usually mutually exclusive , however, there may be good reason for cross-listing between categories
• aim for a balance between breadth and depth; if a hierarchy is too broad, users may have to choose between too many options; if it is too deep, they may have to click too many times to reach the content
• don't restrict yourself to the hierarchical model; some content may be better organised using the database model.

Database models (bottom-up approach)

The database model is useful with relatively homogenous content. Course information and staff directory content are good examples.

In the database model, metadata is the key. Here, the metadata relates to structural elements of content objects. For example unit code, unit name, unit description, prerequisites and year offered are elements of the unit content object.

Linear structure

Use of a linear structure assumes that users will move through the content starting from point A, and progress through an ordered series of page to point Z.

This may be useful for material that requires a linear approach, e.g. some instructional material, but this approach is generally unsuitable for websites with more than a few pages.

Resources

• Information architecture guide
• Evaluating information architecture
• Web Style Guide, 2nd edition
• Card-based classification evaluation

Books

Each of the following is available from the Monash library.

• Louis Rosenfeld and Peter Morville, Information Architecture for the World Wide Web, 2nd edition
• Christina Wodke, Information Architecture: Blueprints for the Web
• Eric L Reiss, Practical Information Architecture: A hands-on approach to structuring successful websites