Information architecture shouldn’t be a big scary thing: it’s simply about creating the same elegance you see in the Golden Gate Bridge or the Eiffel Tower, only instead of being built with steel, it is built with information.

What is Information Architecture?

Information architecture is the process of creating a structure and tools for information such that it can be stored, retrieved, and managed efficiently and effectively. In other words, information architecture is about making information work for you.

Information architecture is different than physical architecture as there aren’t physical materials to arrange. However, the struggle towards effective and simple elegance, which is at the heart of all architecture, has its place in information architecture as well.

When speaking of architecture, we should mention the architect, the person who is responsible. In Greek, the word architect means the chief builder. However, a building architect doesn’t actually build the building. Carpenters and skilled tradesmen do that. An architect, then, is the person who creates the plans, strategies, and direction for the building.

Going back to our case of information, the primary tool the architect uses is “creating meaningful breakdowns”. That is, the architect creates the ability to find information by categorizing it. The following five steps are a straightforward approach to generating your information architecture.

Step #1: Identify Attributes

As humans we use chunking, or treating multiple things like a group, so that we can cope with the sensory and information overload that’s happening all around us. If we want to create groups we need to create attributes and values to group on. The first step is to identify the sea of available attributes. This in and of itself requires that you develop a way to organize the attributes that you identify, because there will be a few.

Identifying the attributes is typically a process of identifying all of the content in your organization that people want to store and find. This might include invoices, purchase orders, time sheets, et cetera. Each of these has a series of attributes such as the invoice date, customer ID, or vendor ID. These may be valuable for organizing the information for retrieval.

One of the cautions in this exercise is that you’ll try to identify every attribute on every piece of content in the organization. While the exercise is in capturing the attributes and the content types, the assumption is that you’ll never be complete and more importantly, you can come back and extend your inventory of attributes later. Your goal is to get the most important attributes down and allow the others to surface if they’re important.

Step #2: Identify Essential Attributes

Once you have an inventory of attributes it’s time to figure out which ones really matter. Attributes of any object or piece of information can be sorted into two categories: essential and accidental. It’s essential that a car has four wheels and that a plane fly: number of wheels and modes of transportation are essential attributes. (No, I don’t count three wheeled vehicles as cars.) The color of the car, say, red and the construction of the airplane, say carbon fiber, are accidental properties.They just happen to be that way.

Essential attributes are used for top levels in a hierarchy. They are at the top of the hierarchy (or hierarchies) because they’re the ones that are easiest for people to identify with.

Step #3: Identify Values

Knowing what the attributes are is a good start, but you still have to get to specific values. For instance, what are the colors? Red, blue, or green?

Are you generally referring to color by its rough category, its Pantone number or its marketing name? What are you going to standardize on for the way that you identify colors? This is a game of standardization and specificity.

Specificity is how specific your values are. Red is a generic color. Pantone 1935C is a specific representation of a red color. If your business is printing, you’ll probably need the specificity of Pantone colors, andif you’re in the business of making wagons, red is probably specific enough.

Step #4: Create Ranges and Groups

Sometimes you can create groups based on the number of values for an attribute. If you have ten color categories that you’re dealing with, the color categories can be your options.

However, if you’re using Pantone colors, which number over 1000, doesn’t make a good set of options for users to choose from in a single menu. To create an information architecture you’ll need to create a set of meaningful breakdowns.

In the colors you may need to create groups based around the primary colors (red, yellow, and blue) or more likely primary and secondary colors (red, orange, yellow, green, blue, and purple). The fun will be in organizing the colors into these groupings since the boundaries will be somewhat arbitrary.

One solution to this: create a poly hierarchy where each color is placed in more than one of these categories. This reduces the specificity (reduction capabilities) of the choice but maximizes the chance that a user will find the color they’re looking for.

Step #5: Design Navigation and Search

The final step with building an information architecture is in connecting it to the visual design. This includes the development of navigation solutions that help the user get to the right information.

The convention of a global navigation across the top and local navigation in the left column is pretty set. Utility navigation, the stuff about how to use the site, is less standardized but also less interesting.

Information architects are most likely to focus on content navigation: helping content authors connect one piece of content with other pieces of content.

Content navigation is built into SharePoint in terms of summary links, but creative use of search can automate some level of the massive task of adding content navigation. For instance, you can add a search web part, which shows all articles authored by the same author, on the same topic, or around the same time.

The design of search itself is mostly about which attributes will be able to be used as facets to refine searches but will include the creation of scopes to create different subsets of the search index in which people can find their answers.

Take it Step by Step

Information architecture doesn’t need to be scary. These steps will put you on the right track to creating an information architecture you can be proud of.

Robert Bogue is a Microsoft MVP for SharePoint, an internationally renowned speaker, and author of 22 books including the SharePoint Shepherd’s Guide for End Users. You can find out more about Robert’s work to encourage business value out of SharePoint at SharePoint Shepherd or more about his technical solutions at Thor Projects.

See all of Robert Bogue's posts on SharePoint usability and SharePoint end-user adoption!