“Out of Date Content Should Be Deleted”

The title of this blog is lifted from Gerry McGovern’s latest New Thinking article ‘How to manage out of date content’; it’s got a real scary example of the damage out of date content can do. Like, a 76% drop in share price for United Airlines after an old news story about bankruptcy from 2002 got picked up as current, scaring the hell out of an already shaky market.

Check out the recent review I did of the research pages of the university. Of 2500 pages, 51% of the pages had not been updated or had no date at all – probably another reason why the experts from the recent research exercise had nothing good to say about the university’s research pages.

And, as Gerry’s article shows, it’s not enough to simply update the page date (either intentionally or not). The content needs to be up to date, to reflect what the person looking at it is trying to do. If that person is wondering what research we did 10 years ago, then fine – let it stay but archived appropriately. But if that person is looking at research we did 10 years ago, but thinks its what we do now, then that’s a big problem.

Research Pages Review or ‘Professors Don’t Read’

Summer’s over and I’m back at my desk – there’s a lot of cool stuff coming up in the Autumn, not least the results of an expert review and usability test I’ve commissioned a consultancy to do of the english web pages. I’m looking forward to it with constructive apprehension – watch this space…

I’ve recently completed a review of the research pages of the university – principally just to find out how many we have…We’ve got at least 2500 pages which directly describe research and there’s probably another 5 000 to 10 000 describing, among other things, researchers, equipment, services etc etc. This review will be used to focus further development work.

One issue I saw immediately, in some cases, is the sheer volume of text. Many pages had at least 300 plus words – which is a problem. Why? Here’s some facts:

- Gerry McGovern, in his book Killer Web Content, says that after the first 300 words you’ve lost 40% of the readers.

or, if you want something a bit more empirical (and we are talking about research, after all)

- Based on the work of Weinreich et al. (2008) Jakob Nielsen showed that most visitors to a web page will look at only 28% of the text (on a page with 300 words, users are on it for around 30 seconds)

Ah wait, but the research pages will be read by Professors and such, who have much more patience than the rest of us time starved hunter gatherers of the web. Sorry:

- A study by the British Library (2008) showed that both undergraduates, and professors, exhibited the same tendency for shallow, ‘flicking’ behaviour in digital libraries. Yep, just like the rest of us, professors don’t read.

So too many words, and you’ve lost the reader. Which is a problem if the most exciting stuff is in your last sentence. And this is now 341 words, which means, according to Gerry, that only about 40% of you ever got this far.

Usability Testing – 15 year review

One of the perks of working at a university, particularly if you’re a geek, is free access to journal articles. When I have time, I try to take a look at what been published recently in the field of usability and Human Computer Interaction.

A recent find is Holingsed and Novick (2007) who provide a review of usability testing practices of the last 15 years. They show that heuristic evaluations and cognitive walkthroughs are the most popular, with formal usability studies being reduced in ambition or abandoned.

I can understand the attractiveness of heuristic testing as its ROI is relatively high. But, one of the reasons I value usability testing, with real volunteers, is that you actually get to meet the customer. It’s often way to easy to avoid the people who use our website and, as a result, avoid tackling some of the major usability issues we may have. Obviously, the method you use must suit the questions you are asking but, for me, every opportunity to meet and understand the customer should be taken.

Holingsed, T. Novick, D. (2007) Usability Inspection Methods after 15 Years of Research and Practice. SIGDOC’07:Proceedings of the 25th annual ACM international conference on Design of communication.

The next SIGDOC conference will be Sept 2008.