I’d like to bring to your attention an article entitled Tinkering With the Virtual Community by Staci D. Kramer, a contributing editor for the Online Journalism Review. She writes on a topic I personally like to file under the category of “Don’t mess with success!“
Ms. Kramer details how a business merger gave need to a web site re-write, in which the authors of the site torqued-off the user base by making a variety of mistakes. For example, “The most visible change was the switch from the original MediaGossip.org domain name to the more staid MediaNews.org. I can understand the name change, but what I can’t understand is why simple (and not so simple) server-based or programatic re-directs weren’t employed? In their defense, MediaNews.org does indeed redirect PoynterOnline.org – though it’d only cost them $8.95 a year to do the same for MediaGossip.org.
However, its not the technological oversights that should concern us. Instead, I’d like you to peruse this article as it brings to light what I consider one of the more egregious sins of site design/redesign – not taking the time to thoroughly engage in a discovery phase. From what I can glean from the article and associated links, it would appear that the redesign effort in question may not have taken full advantage of, or at least didn’t correctly analyzed, the qualitative and quantitative data available to them. Or as Ms. Kramer skillfully details:
The sweeping redesign changed all that – and a lot more. Users lamented the loss of the left-hand column or rail, wailed about the colors and fonts, cringed at the new registration process and, in some cases, hurled epithets at the whole shebang. Some were as cranky as the aforementioned two-year-old when she needs a nap. Others seemed to realize that Bill Mitchell, Online Editor and Marketing Director, and his team at Poynter didn’t set out to make their life hell, that, in fact, they might be on the right track in more ways than not.
Now no one likes change, especially when it comes to software. Still, the entire issue of the removal left-bar navigation does make me wonder whether or not Poynter ponied up the big bucks to observe what users did – versus what they say. Or were the results of such studies overshadowed by their very real and very big technical/database requirements? For more on this point, I would suggest to PoynterOnline.org’s own write-up on the re-write.
Also understand that my intent here is not to publically excoriate Poynter, because to me, at this point in time, the site looks good and navigates well. Instead, I want warn you of a very common pitfall I often find with redeployed church web sites. If nothing else, to help keep your congregation from yelling at you “if it ain’t broke, don’t fix it!”
- Step number one, put down the mouse.
- Step number two, create four or five user profiles that generally cover the demographic expanse of your user base.
- Three, find five people who fit the profiles, and sit them down in front of your current site. Give them a series of tasks, things to find, etc. Have one person aside to see the keystrokes and body language. Another stand behind the monitor to observe the eye and head movement of your test subject.
I know, I know, terribly unscientific here, but unless your church can afford some quality time at the University of Maryland’s Human Computer Interaction Lab… you know where I’m going with this.
- Finally, pour over your user logs. Give special attention to 404′s and other error responses. If you’re awake, the answers should become obvious.
Believe me, I know just how tempting not to take the pains of profiling and testing, especially during a re-write. But if you don’t, you can expect some late nights after the deployment rushing in fixes – which because they’re hurried, may create other problems. Trust me, I’ve been there, I’ve done that, and I don’t ever want to go there again.