"Build it and they will come," said Shoeless Joe's ghost in the movie "Field of Dreams". Without asking questions, Ray Kinsella built his baseball diamond and hoped for the best. Thankfully Ray doesn't have the job of building your company's intranet. Simply put, without knowing who they are, what they need and how they work, Ray would be left standing alone in his cornfield.

Where's the payoff?

Usability has been a buzzword for Web sites for the last few years, and with good reason. Usability and User Centred design methodologies are crucial to the success of all Internet and intranet projects. Studies have shown that if fixing a usability problem is $1 in the discovery phase of your Web project, the cost of fixing that same problem post-implementation will be between $100 and $160!

Skeptics may feel that thorough usability testing takes too much time and resources for the fast-paced need-it-yesterday world of internet/intranet development. I'm telling you it can be done, and to I will show how to do it.

Making it real…

Over a series of five articles, I'll use a real project example to demonstrate how we applied comprehensive Usability theories and methodologies in the development of the information architecture for an intranet. This series of articles will be broken down into five segments:

• Understanding Your Users: Stakeholder Analysis,
• Defining Requirements: User Needs Analysis (UNA),
• Developing a Concept: HTML Prototype Development
• Testing the Concept: Confirmation Testing

Understanding your users

Our client is an educational institution with a population of over 20,000 students and four campuses. The goal was to develop an information architecture for an intranet that would serve the employees and, indirectly, the students.

Analyzing the stakeholders

Our first method of gathering data for the project was a stakeholder analysis. The purpose of the stakeholder analysis was to develop a better understanding for project expectations and scope, to learn more about the organization, and to gain a general understanding about the end users.

To develop our stakeholder analysis, we gathered information by meeting with the organization's intranet steering committee and conducting a series of interviews.

The intranet steering committee was decided upon by the organization and comprised of three members of the college's Tech Department and two members of the Public Affairs department. Our initial meetings were useful in generating basic organizational information. For in-depth information on organizational structure, stakeholder expectations and current means of organizational communications, we turned to stakeholder interviews.

Determine the approach

Through discussions and meetings with the project steering committee we decided to conduct individual one-hour tape-recorded interviews with the president, vice-presidents and selected directors of the organization. At this exploratory phase, we decided to use "unstructured interviews to allow our subjects to lead the discussion. We wanted the interview subject to lead us to in the discussions. We decided against focus groups because it would have been difficult to get all of the stakeholders together. Also interpersonal dynamics and power structure might have restricted people from giving their opinions, thus skewing our data.

As a prelude to our interviews we provided all participants with an overview of the project and a few general questions that we would be discussing. Here's a sample of the questions we posed (there were 10 questions in all):

• How does your division interact with the rest of the college.
• What is your relationship to the college's various sectors and departments?
• What are the most important pieces of information that these departments need to function properly?

We found that this step was useful because the interview subjects were well prepared.

Getting the Results

From the interviews, we developed an understanding for the college's organizational structure and established three categories of end users that we would interview in the needs analysis phase: faculty, support staff and administration.

It also became clear that the dynamic between the college's main campus, and its three satellite campuses needed consideration in the upcoming user needs analysis phase.

Apart from our own information gathering process, the interviews educated the stakeholders as to the nature of an intranet and its' implications on the communication structure of the organization.

Also important, but more subtle, these interviews gave us a reading on the "political" component of the project. Understanding organizational politics is an important element in the success of an intranet. Initial support at the high levels and key positions in the organization must be built in to help the intranet project to succeed. A barometer reading on how ready staff is to adopt an intranet and identifying those who are resistant to it is important intelligence for your client.

Some Lessons Learned

Using unstructured interviews as our primary data gathering methodology for the stakeholder analysis meant spending one to two hours with each interview subject. There would have been cost benefits in holding a focus group with all of the stakeholders present because the session would have only taken roughly two hours to complete rather than the ten hours we actually spent on individual interviews. This focus group session could have been combined with follow-up telephone interviews to fill in any information gaps.

However, we feel the interviews generated richer data than a focus group would have. In addition we could never be sure with focus groups how interpersonal dynamics would impact on participants' responses. For example, would a director feel free to speak about his or her communication problems in the presence of the president of the organization? Also problematic would be getting all of the stakeholders together at the same time and place. The opportunity for people to discuss and debate ideas in a focus group might have flushed out information, which we missed in the interviews. Given a larger budget and longer timeline, we would have considered one or two focus groups in combination with the interviews we did conduct.