Making Cents from Information Architecture

In: Columns > The $ & Sense of IT

By Alan K’necht

Published on January 22, 2003

When it comes to Web development, everybody has taken short cuts over the years. This holds especially true when working on low budget projects. One of the most costly short cuts is skipping the development of a sound and highly functional information architecture (IA). While this short cut may take several forms, failure to devote enough resources and to document it properly will cost the owner of the Web site more than just a few cents.

Why do these short cuts happen? Many developers think they know what IA is and do it in their heads. Some do know what IA is, but still do it in their heads. It might work for a three-page Web site, but what happens when the site grows, or when you need to share the IA with another Web developer, or when the Web site has more content or new features that need to be added? Additional costs pile up very quickly. To paraphrase a famous TV commercial, “You can pay me now or you can pay me more later.”

First, the hard part: defining what IA is and isn’t. A quick search of the Web will unearth thousands of definitions for IA. Some seem to be saying the same thing, while others say something completely different–whether too vague or too specific.

One definition I like comes from Mattie Langenberg:

“Information architecture, as the name implies, is basically about taking content, and a structure to present that content, to an audience. It is the information architect’s job to ensure that information is well organized and presented in an easily accessible interface.”

I’ve personally used a more streamlined one when explaining it to clients:

“Information architecture is a combination of art and science to organize information into a functional and usable format, allowing someone unfamiliar with the information to easily find what they’re looking for.”

One of the most universally recognized instances of IA is the public library. I’m not talking of the public library of today with computer-based searches on more categories than imaginable, but the library of a few years ago–the library with rows and rows of index cards.

Libraries that use card indices may seem archaic, but they are actually very modern. Long gone are the days of private libraries where books were filed by the preference of the owner. Perhaps it was by the author’s name, the title, the subject matter, or even the nature of the publication (play versus novel). A visitor to one these libraries couldn’t be assured of finding what they wanted without the help of the librarian who organized it. Modern libraries quickly adopted something called the Dewey Decimal System (DDS), developed by the librarian Melvil Dewey in 1876. His concept of dividing books into one of ten categories revolutionized the library world. The DDS was the first step in creating a highly organized and flexible IA schema for books.

Librarians took DDS as a foundation and went on to develop a highly complex system of organizing books that was easy to use by anyone. For every book in the library, the librarian assigned a number to it (according to the DDS) and created several index cards–one for title, one for author’s name, and others as required. Each card contained the same information but was indexed by different key fields. A user simply needed to know the name of the author or the book to find its location in the library. These index cards could pinpoint a single book from tens of thousands.

This concept of IA worked wonders if you knew the book’s name or author. How about for general research? An index for the DDS organized information in primary categories (1 out of 10) and then broke those down into smaller and smaller units, so the user knew where the books on a particular subject were located. This system, developed long before the invention of computers, has no problem handling all the books now published on the subjects of computers and computer programs.

This forethought and flexibility is what’s needed when addressing the issue of IA and Web site development. Imagine what it would have cost each library if the DDS couldn’t have handled all the new topics and subtopics to come in the last 125 years. How many times would they have had to reorganize their systems and redesign their card indices. This may sound silly, but how many times have you had to rethink how information was organized for a Web site or needed to squeeze one more heading into a navigation bar?

All these redesigns, while good for generating long-term employment, don’t leave a good taste in the mouths of those who write the checks. The time and effort in developing a sound IA for each Web site needs to become a priority, just as a librarian would do in organizing the contents of a new library.

With this understanding about IA, what is its value in Web development? Following the library example, IA is not what the library looks like, nor what the shelves are made of, nor where they are located within the library. IA is how the books are organized. The index cards are the efficient navigational tools that help users find what they’re looking for.

Organizing the data should be the first thing done in Web development (catalogue those books), but all too often designers start by working on the visual appearance of the site–what colors to use, where to put the navigation, etc. How can a designer decide where to put the navigation when what is being navigated isn’t defined and the information isn’t organized?

Web developers need to remember that IA is the skeleton of a Web site, the business logic (database queries, javascripts, cgi, etc.) the muscles, and the HTML code the skin. Another analogy might be building a house. In this instance, the IA is the blueprint from which everything develops. Your database is the foundation, the data queries the frame, and the HTML is simply the drywall and paint. Without the architect’s plan, the house might not be functional.

So why do many Web developers start with a look and feel concept? This is the question that needs to be answered by developers and explained to clients. Many people think the artist’s first renderings of the site’s different pages is information architecture. They are not; they represent presentation architecture, how information gathered from the IA should be displayed. This lends itself to usability. Going back to the earlier library example, the presentation architecture is simply the design of the index cards. The information on the cards (Web page) needs to be easy to find and easy to read.

When justifying the time and money (the return on investment–ROI) required to properly investigate, document, and develop IA, many business people and Information Architects themselves talk in the grand scheme of things and the bigger picture. Rarely do they identify savings and benefits. Since IA is seen merely as a cost, and the look and feel gets all the glamour, it’s no wonder IA gets pushed back or, even worse, gets dropped entirely from the project plan. Some of the benefits of IA, to create a positive ROI for the suits of this world, are:

Search Engine placement – By organizing the Web site into appropriate sections (with corresponding titles), IA will identify up front some of the key elements that search engines use to help rank pages (elements include page name, key words, directory names, and file names). What is it worth to a company to have a Web page listed in the top 10 or 20 search results?

Improved usability – When proper IA is applied to Web development, it results in a more usable Web site. This is not to say that a Web site design doesn’t have to be evaluated for usability. The key navigational elements will be identified and organized, making the job of the graphic artist easier and quicker. Think about the library’s index cards. In every library you can locate a book by multiple factors (author, title, etc.), so why should Web site navigation stick information only under one heading. IA provides these categories.

Locating Files – When it comes to maintaining large Web sites, think about how long it takes to locate a specific file that needs updating. The better the IA, the quicker you can find them. Good IA identifies the categories appropriate to the information. As a Web developer, you should follow these guidelines when creating the directory structure and file names. Then, when a change request comes in, you no longer have to go to the site and search for the page just to find the URL to be able to locate the file name. You can now zoom in on the file from your favorite editor. Remember, the naming structure harvested from the IA (directory name and file name) is also used by search engines to help determine page rankings (see point 1). How much will this reduce future development costs? That is hard to define, but easily can range from 1-5%, depending on the frequency and the nature of changes.

No Need To Redesign – How many times have you been faced with redesigning major portions of a Web site simply because something new has been added. With IA, just like the Dewey Decimal System, you should be able to simply slip in the additional content without having to redesign the entire site. How much can this save over the course of a couple of years? Thousands upon thousands of dollars, depending on the size and complexity of the site.

Reduced Development Time – By taking the time up front to develop a proper and signed-off IA for a Web site, development time can easily be reduced by 10-25%.

Information Architecture can be something definable, explainable, and valuable to the art of Web development. Web developers need to embrace IA and push for it on all their projects. If you’re ever in need of a great Information Architect, look no further than the public library and anyone who has a degree in Library Science.

• Add to Magnolia
• Add to My Yahoo!
• Add to Newsvine
• Digg this story
• Add to Del.icio.us

Related Topics: Information Architecture, ROI