George Mason University: Web Page Redesign Issues
New title? This one is kind of bland.
Go to Earlier Versions
Redesigning a website is always a challenge, as we can attest after taking on a major redesign of the George Mason University website in February 1998. Because the GMU umbrella supports more than 500 second-level sites from the top [index?] page, our primary goal was to enable users to find information effectively and efficiently. Our secondary goals evolved out of Mason's financial and marketing concerns: because resources were an issue for the University, we aimed to develop a site with easy content management and update; and because previous designs were either outdated or conceived as temporary highlights, we wanted the site to consistently present a professional, high-tech image of the university.
We did not rely on any one organization to complete the redesign. Instead, we coordinated a team comprised of individuals from across the University, each of whom brought different experience and skill sets to the effort. University Libraries, University Relations, University Computing & Information Systems (UCIS), and Institutional Research and Reporting were all represented on the redesign team. [Did this team have an official name?] Next, we [meaning the team?] performed a survey of other university websites. We felt that it was important to identify, first, what other higher-education institutions were doing with their top-level pages, and second, what elements (if any) were consistent on the websites. While we wanted to address our audience in the most effective manner possible, we also wanted to provide links that are familiar to users.
In our research, we found that most university websites operate via "one-way" navigation. This type of navigation requires users to access secondary pages through selected categories (i.e., a user finds a link for the Physics Department homepage only after clicking on the category "Academics"); most of the secondary pages are associated with only one category. Some of the most popular categories are: Academics/Academic Life, Admissions, Arts/Sports, Libraries/Computing, News/Events, and Info/Weather. Many of the sites we looked at also include a search box that allows users to type in keywords and receive a result set of pages containing those words.
Both versions of our websiteour "original" version (http://www.gmu.edu/webdev98/index1.html) and our "temporary" version (http://www.gmu.edu/webdev98/index2.html)also used one-way navigation; consequently, each of our secondary sites was available through only one category on the index page. We knew that this system was frustrating to users, many of whom considered it a challenge to find something as simple as the Human Resources page. Because the redesign team's primary goal was to provide users with an uncomplicated and effective way to access all secondary pages, we decided that the one-way approach to navigation required serious modification. Mason's redesigned site (http://www.gmu.edu/) therefore incorporates several different navigation options so that, from the very first page, users are able to navigate the site in the ways that feel most comfortable to them.
Maintenance and Technology
Though we are excited about the flexibility of the redesigned site, it does require a significant amount of maintenance. Both the "original" and "temporary" sites required oversight on approximately 7 to 10 pages. Our multiple navigational structure increases that number to more than 35 pages. To accommodate this number of pages without an increase in resources, the team implemented what is called "database backend technology."
In short, this database technology organizes all the lower-level links and directs their delivery to Mason's top pages. Each website within the Mason umbrella has its own entry in the database. For each site, fields in the database entry specify the creation date, alphabetical listing(s), audience relevance, and categorical distinctionas well as contact and other administrative information. The database technology also provides a content-management interface and link-delivery system for the news and events portion of the Mason site, Today@Mason. Today@Mason is updated and maintained by University Relations rather than the Web team; as a result, it resides on a different system. With the database technology, University Relations content providers have the interface they need to enter daily news stories.
Cold Fusion technology makes it possible for the team to administer all the information in the database from a web interface. Selections and modifications to the database allow the team to quickly update administrative information as well as link-delivery informationwhich audience pages and which category pages links appear on, for exampleas well as the order in which they appear and how they are grouped on each of the pages. The integrity of the data is maintained because it is contained only once in the database; the information is distributed to, however, to as many sites within the Mason top pages as is appropriate.
The Mason website database and the Today@Mason database provide two significant functions: First, they allow URLs to be displayed on as many pages as is appropriate even though the information is maintained in only one source file. Thus, once an update is made to the database, that change is reflected on all the associated pages. Second, database technology allows content providers (in Mason's case, the redesign team members and the University Relations staff) to directly add information and/or content to the site. No emailing or calling a webmasteror waiting for a responseis necessary. Providers make the changes they require, when they require them, with no intermediate steps.
The major portion of the redesign spanned a 3-month period and involved 3 full-time designers/developers as well as various management and part-time staff. Clearly one of the strengths of Mason's approach to the project was the establishment of a cross-departmental design team. Too often, design efforts are centered in one department; that kind of narrow team vision results in a site that is limited in its usefulness, that is "lopsided" in one fashion or another. We believe that Mason's new site bears no such flaws, for not only were the departmental associations of the team members varied, so were their skill sets. Individual members provided expertise in such areas as design, development, system administration, and management. It is this mix of associations and skill sets that brought success to the Mason effort.
Working with focus groups also contributed to the development of an effective usability scale [what is an "effective usability scale"?]. Focus group and user comments provided immediate guidance on potential problems and much-needed input about the links and information on audience pages. However, the redesign team found it necessary to carefully scrutinize focus-group data for four reasons: first, bias was inherent in the feedback provided by individual commenters; second, the focus groups rarely came to a consensus regarding the pages; third, the groups often debated issues of design rather than issues of content; and fourth, no focus group looked at the overall site. Because each group concentrated on a relatively small portion of the site, it was the redesign team's responsibility to take the groups' input, put it in context, and consider it when making strategic decisions about the overall design of the site.
Our experiences indicate that completing the redesign is only the tip of the iceberg. The development of an information-management system and university marketing tool requires continual analysis and refinement. The administrators of top-level university websites should have long-term goals which will set the standard for consistent navigational approaches, information organization, design elements, and content management throughout secondary pages. They should also have a non-biased tool for measuring which changes to the site are proving useful and which could benefit from continued rethinking. (So far Mason has no such means of assessing the redesigned site's effectiveness.) Keeping any site fresh and alive requires a united university effort, a dedicated Web development team with a vision of the future and, as in our case, a vision of the university website environment as a whole.
[The conclusion isn't all that strong. Do you really have no statistics about the number of hits the new site (v. the old site) is getting? Anything stronger that you could add? Comments from users? Has any kind of feedback page been set up?]