No reboot on May 1st for me
I will be missing the May 1st CSS reboot this year. Time and life conspired against me and instead of putting up a half-arsed redesign I will spend a couple weeks and make something worth while. Some rebooters sites are starting to appear tonight and there are some really nice designs out there. Some sites are broken as people switch over (including the CSS reboot site at the time I posted this), let the chaos begin! 
Know your audience, not your web site
Chatting with Derek and venting about issues that have come up in the day-to-day related to log file and traffic analysis, I have come up with my catch phrase for that pretty much sums up my thoughts on the whole thing with regards to web development/design/etc: know your audience, not your web site. This new ‘mantra’ of mine comes from a few discussions recently that are motivated this new found obsession with web stats that seems to pop up (maybe thanks to Mint,Measuremap, and Analytics ) and studying user traffic on your web site.
The intent of the whole know your audience, not your web site thought is to be a reminder about the importance of the audience over the numbers found on the server. It is my mantra for the summer.
Know your audience
A quick search for that phrase on Google will give you a load of results. Knowing your audience is the key to any presentation, written or multimedia or in person. Getting to know your audience can be achieved using a number of different approaches from straight assumption based on anecdote to observational study (usability studies). Does the value of this really need to be explained? It seems like a no-brainer besides the little problem of how do you get to know them.
The answer appears to be simple: watch the site traffic patterns and see what content the users are going to. If we get real tricky with something like Analytics or Mint we can look at how long they have been on that page too. What does that actually tell you about your audience? Not much I think. It tells you more about your web site than anything else. If you look at your page and see that people are going to this one page a lot but are spending very little time there it could be that:
- Your content has a popular keyword in it that ranks high in search results so people go to that page, scan it, move on.
- Your content is well written and people just scan and get what they need.
- The page sucks and people think they have the right place but the page sucks so much they just leave.
- I could go on.
The point is you really have no idea why people are leaving your page. They could be happy with it, hate it, or not even looking for that page. You don’t know unless you know what audience that page is supposed to cater too and what that audience is looking for.
Know not your web site
I am by no means saying that log files and traffic analysis should be ignored. What I am saying is that the emphasis should be on the audience and what they want first, your web site should be studied only once you have a good grasp on who these people are. Traffic analysis and assessment of value with regards to web content as a key part of a pro-active approach to your web site content management is essentially voodoo.
The value of this data comes from being able to study patterns of usage over time and figure out if a site you have put a lot of resources into generates the traffic to justify the expense. If it doesn’t maybe you need to ask your audience what is wrong? It may be nothing and you really should ask.
“I want to know my audience”
First you need to identify your audiences. Pick your primary, secondary, tertiary audiences – sure you can have more than one audience in those categories but it is critical you identify them. Next you want to solicit feedback from those audiences and assign weight to those responses relative to their status as primary, secondary, or tertiary. Just be sure to get feedback from more than just a few otherwise your information could be a little skewed one way or another.
You don’t need too do it but a usability study is invaluable if you can get enough participants from your primary audience at least. If you have a diverse primary audience like say the UW home page than try to get a couple from each group. I will post more about how to do a test at UW in the future. Really, just make the effort to get to identify and get to know them as your audiences will appreciate it and from that you can better utilize your server data.
This is my whole motivation around doing the usability study of the home page at least once a year. I want to observe people using the page, hear from them their thoughts, see their reactions to links and images. From a small group of them I can start to see patterns and hopefully identify things that need to fixed. Also, I hope to be able to better anticipate features that might make their experience better. I want to you, the audience, as best as I can.
Durham College redesign is looking for feedback on current (old) design
If you have a few minutes take a look at Durham College’s web site and then drop by the design blog and tell them what you think—of their current site.I am not entirely sure if they have a before and after shot but maybe you nice folks that read this blog can bug them about that They are looking for feedback on their current design and what can be improved. It might be especially valuable for students and others that have no idea what or where Durham College is to give them feedback.
Speaking of which I am pretty much out of time to pretty this place up for May 1st’s reboot…
update: need to learn to read email better, I have fixed the wording to reflect that they are looking for feedback on their current site that hasn’t been redesigned yet.
Usability data making me crazy
It could be the fact that I just hate writing reports but pulling together the data from the testing we did in the fall isn’t as fun as I thought it would be. For internal audiences we ended up with 24 participants, 12 students and 12 faculty and staff. What stands out right now is how bizarre all you people are
External audiences don’t you worry. You will be tested, just wait. Once this report is done the focus will be on devising an effective way to automate this testing and allow you to do it from home for a nice little reward.
You could be asking why, in April, am I just getting at the data now and I should have a good excuse. But I don’t. The motivation now is getting this report done so all the time I will be spending on the 50th Anniversary won’t mean this report is never finished. Expect to see a draft report the end of next week.
IE 7 list-item whitespace issue: not a big deal but annoying
Inspired by Veerle’s post on the issue to actually fix this problem on the UW home page and IE 7, I went through and applied a simple fix to the CSS that is in the custom style sheet to the home page. This is what you get if you use the holly hack to fix IE 6’s whitespace issues because IE 7 ignores the holly hack but didn’t fix this bug… What I started with is a menu that looked like:
With the CSS:#secondarycontent li { font-size: 80%; width: 100%;}
…and changed it to:
#secondarycontent li a { font-size: 80%; width: 155px;}
So now it looks like:
Applying a width to ‘li a’ (not just li, it doesn’t work) is all I needed to do. I likely would have saved myself some time if I did it properly in the first place. Oh and its IE 7’s improvements that make it ignore the fix for IE 6… is that not swell?
