good health-related sites

Krug mentions some Web awards--I bet they'd be a good reference. I'll see if they have any winners in this category.

Don't make me think (about cancer)

It look me a long time to get around to reading it, but very little time to actually get through Don't Make Me Think which is a terrific endorsement of a usability book, really. And it was really helpful! It was all interesting, but I think that what will be most helpful for this project is his dos and don't of page layout and navigation--I will review the chapters in the polar bear book for more information, too, because I think that herein lies a lot of the problem with our site. People tend to get lost in it.

Some of what Krug says is at odds with the purpose of health-related info on our site, which is interesting. He talks about taking out words, but we HAVE to have them--it is all very heavy on information. How do we know at what point people will just stop looking? I guess that is related to another of his points, which really hit home: you build up a reserve of goodwill (or anger) as you use a Web site, and your trust level that the info will be there, for you to find, and that it will be accurate, often determines frustration level. So we shouldn't hide the contact info (according to Krug) and we need to be sure and minimize the happy talk (there's quite a bit of this about the good we do, blah blah)--and just get people the information they want. And not try to keep hammering the grantee aspects. Seems like a delicate balance. It would be helpful to find a Web site that does this really well-- all his examples are commercial, and I haven't seen a Komen site that really gets it right.

The final and probably most important point that got across to me in Krug's book is the absolute necessity of usability testing, I think it would be inexcusable to not do at least 1 round--I'll email Dr Bias and figure out what the deal is with the Lab. I believe Krug's argument that even slapdash usability testing is still 100 times better than none at all, so, really, it just has to happen. Plus I've never done it before, I think it will be fun (sort of). In a grad school kind of way.

user informational needs

I met with Mary Moore last week and it looks like the educational materials and contact info on the Web site are, for the most part, up to date. She went over for me the kinds of questions she encounters, what kind of info people are commonly looking for, and what the process usually is for them to find the answers to typical questions. The upshot of it all is that most of the answers are in fact on the site, but people don't understand how to find them, and often they also don't understand what the terminology means. For example, people often need breast screenings, not mammograms, but they don't know what screenings are and thus can't find the contact info on the site--we decided to get around this by setting up a button that says, "Free Mammograms" and pulling people in that way, to explain the difference later, rather that trying to educate them first.

I showed Mary Moore the map on the Houston Web site and we agreed that a simple Google maps function is all that we really need in Austin. We like the idea of the form they use to sort, but Mary Moore says that since everything is county-specific, we need only forms that have a dropdown list of services in each county. I should have enough PHP to build this, from the info on the site in other places and some additional resources that she gave me.

Another concern is the rapidity of change in these services, especially out in the outer counties, but I think that planning to update the site at each grant period will be a good way to ensure that someone reviews all of the info at least twice a year. Additionally, since Komen funds the majority of the services, we are going to limit the services listed (where possible) to the grantees, so that we can guarantee their quality.

We also discussed forms to capture more information about how educational resources are used. Because people tend to call or email her directly for scheduling presentations and health fairs, Komen has no way to track the distribution or the cost of materials given out. Removing her contact info and replacing it with a form seems like a good idea, but I am ambivalent about deliberately making it more difficult for people to reach us. I may take this to Ramona, and/or see if there's a less annoying way to capture the info Mary Moore wants.

Finally, Mary Moore also stressed the need for it to be as clear as possible that Komen is the engine behind many of these services, and to relate all this to the Race for the Cure. The Race raises the money to fund all of these clinics and hospital departments, but now on the site there are no crosswalks from Race to Grantee info (such as how to apply) to info for individuals who need services. It will be a challenge to develop these without getting people hopelessly lost! Or rebuilding too many pages. I hope to have some outlines for some of the indivigual pages done soon, to get Mary Moore's input. I also bought the IA polar bear book and am referring to it for guidance on how to set up the navigation to try and link these multiple concepts without further confusing everybody.

meeting scheduled

I have scheduled a meeting for this weekend with Mary Moore Cavanagh, the Director of Education and Outreach at Komen Austin. I want to get her feedback about the kinds of health questions people have when they call her, especially since they do not seem to seek information over the Web--not that it's on our Web site anyway. I wonder if the kinds of things they want to know are more comfortably asked of a person? or perhaps they know from previous experience what kinds of info are on the site? or, since many of the questions we get are from women looking for free care, perhaps they do not have internet access and got the phone number from some other source. I am interested in both who contacts us, and what kinds of info would be most helpful for them. Mary Moore has been at Komen for years, and also talks to other Affiliates and travels to the national office in Dallas on a regular basis, so she should be very helpful.

Web usage logs

I found the usage logs on Convio and have begun to analyze them. Some of the data within might be useful, if I can provide context. I was looking to see if anything jumped out on the educational pages, how usage compared in race/nonrace season (who looks us up in Feb, the slowest month? and why?), and how people arrive/depart the site. I want to know if they're looking for educational info and if they found useful information. The available stats, broken down by month, include:

• time of visits--41% under 30 secs in Sept! even higher in the non-race season-- 70% in Feb
• pages viewed: in Feb, race info (even at 9 months away), some participatory stuff like volunteer info and calendars, contact us, then eventually education/outreach, below merchandise. In Feb 175 views total, of 4600 visits.
• people look at work--all usage is lower on the weekends.for both seasons.
  
• search: people use the name, not general queries. they already know of our existence, I guess. Breast cancer Austin will still get you there, though.
  
• lots of people leave after finding the contact info--that page is 6th on the exit list in Feb. In Sept, race season, it moves down to around 20th. At least the race pieces generally work well, and better than they did last year it seems, if I can make the assumption: contact us was the exit page 900 times in Sept 97, 770 in 2008.
  
• entry: in Sept, from race donation pages, from the National site on the find-a-race page, and from local sponsors and big teams. In Feb, they came from the national site (find an affiliate), from volunteer stuff--I did not realize that was so important, although many of those may be repeat visitors--and local media outlets. NONE of them came from health resource sites.

I picked up Don't make me think, the book Dr Bias uses in his usability class, to try and give myself some background on what this stuff might mean and what we need to do to improve it. I will report on that in the next couple of days. There may be more reports that could be helpful also, I'll keep poking around Convio.