So why do they tend to suck so badly?
Here’s my Twitter-sized response:
Most uni websites suck because they’re designed by committees who are unaware of how the web is consumed. Or they simply don’t care.
Needless to say, this is a condensed (and admittedly sharp) version of my views. Of course, there are many factors involved, notably resource limitations and red tape. But from the lens of results-based analysis, it indicates that proper web development/design practices are not valued highly enough.
As someone who has transitioned from being a student to an employee at the same university, if I were tasked with constructing a web presence for a university, this is what I’d look to do:
Settle on a Design Language
(But Leave Room for Sensible Rule-breaking)
By ‘design language’, I’m referring to the set of visual elements tied together to represent a brand. The aim is to achieve a unified visual consistency across all publishings or products to identify the brand.
This means establishing a branding and identity guideline. One that includes well-documented and continually-updated building blocks like:
- image, code and other asset libraries
- usage examples
- writing style guides
- decision rationale
In my opinion, it is also quite important that publishings are visually beautiful. Details like professional typefaces (through tools like Typekit), subtle animations, text aliasing, interactivity, touch-based features (e.g. swipability) all really add up for a great user experience and positively affect the brand. Here are some websites that demonstrate this well:
On Flexibility and Rule-breaking
A university is a diverse collection of organisations. Each organisation within the university and their website(s) may have requirements that are important and unique. For example, say the Division of Marine Studies would like to use the colour blue (for obvious reasons). If the university forces the use of red, because red is the official colour of the university, then we have an issue. Imagine that the website’s primary aim is to sell the Marine Science degree: would having red splashed all over the page be necessarily effective?
Rather than stifling needs, simply have a minimal set of unifying elements that identify a website as belonging to the university. Including the university’s logo may be enough to achieve this and other elements may differ as appropriate (e.g. different layout, colours, navigation, background). Let each site’s personality show!
How exactly do we achieve this? How do we coordinate the various oft-isolated organisational units into applying these guidelines appropriately?
Open Up Communication Channels
Set Up an Intranet
Discounting physical meetings, any kind of social network (e.g. Yammer, Podio) or wiki (e.g. Atlassian’s Confluence) would be a good starting point. Somewhat obviously, this is just a way to disseminate information but also to provide the general community within the institution a chance to influence the design decisions being made. At the very least, it’ll serve to increase awareness of what’s happening.
Through this intranet, anything of significance should be being documented sufficiently, if not comprehensively. The branding and design guideline especially should be given plenty of attention and crafting.
Here’s an example of the intranet that Microsoft uses: http://www.redoctober.com/content/detail/microsoft_corporate_intranet_redesign/
Foster an Open Source Culture
Anything with your name on it that is publicly available bears with it a reputation component. People care about reputation, and that is a strong driver for people to produce quality work — whether it be code, artwork, writing, whatever.
Even if the projects are not open sourced to the public, there is still much to be gained by opening it up to the members of the institution. There will likely be many who are interested in making improvements and reporting issues. The key aim of this is to empower people. Again, this fits well into the walled garden that is an intranet.
Easy Feedback Submission
Related to the previous topic, people — especially front-end users — need be able to submit their thoughts and issues easily. To this end, use low-resistance tools such as:
It’s imperative that the feedback process takes a minimal amount of effort. This takes advantage of the impulsive nature of users — this is partly why the Facebook ‘Like’ button is so popular.
It is also important to make sure people are being listened to. You may not implement a suggestion, but take the time explain to the person who took the time to submit feedback why such a decision was made.
Hire Supergood People
To make the organisation into a outlet for great quality work, you need to have a super engine. An organisation, by definition, is about people, so hire damn good ones.
Hire Enough People
Sure, you can have the right people, but you may not have enough of them. So hire enough and also give them the quality tools to do their job.
Establish an Audience and Target Away
Universities cater for many audiences. These are the main audiences I see:
- Students (obviously)
- Current students
- Potential students
- Past students (alumni)
- Staff members and staff-like members (e.g. postgraduate students)
- Teachers (academics, tutors, practical demonstrators)
- Academics and researchers
- General staff members
- Potential staff members (including researchers)
- Industry, business and community members
Notice these categories aren’t unique to university websites either. In broad terms, you’ll have two main divisions with many (if not most) wesbites: (1) consumers already familiar with your website/product/service and (2) potential consumers. Both need to be properly catered for.
This lack of proper catering, is precisely where many university websites fall over. The typical problem I see is that a given webpage tries to cater for more than one audience ineffectively or for an inappropriate type of audience(s). The target audience for any particular page ought to be clear and well-planned.
Technology changes quickly and information is outdated quickly. Organisations must learn to surf this techonological torrent of titanic proportions. As a result, change must be ready to be made and also easily made — but only when new evidence emerges. The web is ephemeral. It is not a print publication. Whilst some care should be taken, overcautiousness and paralysis-through-analysis should be avoided.
Test and Test Some More
How can you tell whether a website is being effective? Conduct on-going testing, of course! This includes:
- Hallway usability testing (with the right audience members)
- Actively listening to feedback (see ‘Easy Feedback Submission’ section above)
- Traffic analysis (e.g. via Google Analytics)
- Automated browser testing (e.g. via Selenium)
- Browser compatibility testing (check out Adobe BrowserLab and BrowserStack)
- Server monitoring (a la New Relic and Sentry)
That’s about all I have for now. Will likely tighten the writing up and add things over time.