University Website - xkcd.com cartoon by Randall Monroe

Every university home page is a delicately judged balancing act.  As the main door into a large, complex organisation, the home page must serve many different audiences — prospective students and their parents, alumni, professional staff, academic staff, current students at all levels, the media/marketing department, journalists, visiting academics, employers, sponsors and investors, government agencies, professional associations, benefactors, random members of the public who heard about a public lecture they’d like to attend and it starts in, like, 15 minutes from now, could you just tell me where it’s happening?

Claire Spencer and I presented a poster at Ausweb 2005 on emerging best-practice for university home pages. The poster was based on a competitor research project comparing 68 university web sites.

In turn, the competitor research was part of a larger project aimed at making some evidence-based decisions about the design of a new home page for the large research university that employed us.

We undertook several types of research:

• user research — an online survey about preferences, ideas and opinions
• business requirements analysis — a survey of internal stakeholders, asking them to prioritise different types of information and online service; and a review of corporate strategies, plans and other documentation about business goals (changes in the student enrolment profile, increased openness about research activity and outputs, key marketing messages, etc)
• user personas — task analysis, identification of key user groups and characteristics
• usability assessment – task-based testing in a usability laboratory, expert walk-through of our web site and several others
• web standards – identifying the coding, accessibility, performance and related standards we wanted to use in the new design templates
• content analysis – extensive review of the content and services already available on the University’s many, many (many!) web sites and in its publications (brochures, newsletters etc)
• site traffic analysis – understanding which parts of the University web site are important or heavily-used, and who uses them (or should use them)
• search log analysis – identifying sought-after content, understanding the language our web users employ (so we can use it in our web content); by undertaking ‘test searches’ we gained an understanding of why users preferred search over link-clicking for some types of content
• stakeholder consultation – interviews with senior managers, communication and marketing officers, content owners, subject matter experts and other internal stakeholders, to understand their ‘pain points’ and identify how the web site might help to resolve these

This research took about three months of intensive work by Claire and me, with help from other members of the project team. At times our manager and some colleagues chafed about the effort that went into this research phase.

However, when it was all synthesised into a design brief and proposed site structure, the value of the preparatory work became clear.

An immediate benefit was that the development stage was relatively fast and simple. Everyone knew what was required and what technical standards we needed to meet. We had a clear task list and timeline. Testing of the prototypes was mainly technical — validating code, checking download times,  and so on. Some lightweight user testing confirmed that we were on the right track with the navigation and visual design.

Another benefit of doing the groundwork emerged when we reached the ‘final approval’ stage for the new design. Whereas previous home-page designs had been the subject of months of debate before being signed off (if they were signed off at all), this time the proposed design was almost immediately endorsed by the relevant authorities, with only one or two minor changes requested.

[grins] Did I say “relevant authorities”? Don’t let that term mislead you: our web governance model at the time was nothing to be proud of. Several senior executives — and quite a few of their direct reports — each thought they had the final say over what went onto the home page. Even a minor change to the wording of a link could lead to months of argument, competition and resentment.

For the web team, tucked into a low-profile niche quite a long way down the organisational hierarchy, the only way to resolve disputes was by careful consultation and negotiation with the higher-ups, often at arm’s length (eg our manager’s boss might conduct the negotiations on our behalf, instead of letting the project manager speak for herself). We had no power and very little access to senior decision-makers. Instead of trying to acquire formal authority ourselves, we took a dual approach that established our credibility.

First, as outlined above, we invested time in ensuring our decisions were soundly evidence-based — an absolute must in large research-oriented university where every staff member believes they are entitled to question every decision.

Secondly, we worked on developing relationships with people who had an interest in the web and could advocate on our behalf. We provided a range of services to faculties and departments, including a broad-ranging training program, professional support via a community of practice, and a consultancy service to help them undertake their own web redevelopment projects. Business managers, communication specialists, IT managers and web staff all benefited from these services, and in turn they provided lobbying support for the redesign project and other University-wide initiatives related to web management.

.

Bonus: Gregory Beyer’s review of a non-existent book, “The Devil and the Rising Sun: A Year Inside the West Carolina University Admissions Department” by Cortoroy Chen, makes me wish it existed. Very funny, if you’ve ever worked at a university.

And a non-humorous bonus: web managers and content strategists will find lots of useful standards, frameworks, checklists and other tools on the Victorian eGovernment web site.

You might also be interested in...

• Three ways to deliver intelligence data (0)
• What futurists do (0)
• Toolkit: SnagIt (0)
• Toolkit: Simple Tags plugin for WordPress (0)