Welcome!
other pages
Basic Skills | The
Demographic Environment | Web site critique
Marketing site critique | Go Shopping | Marketing
Site Features
Web page critique form
Stewardship page critique form
New! In the Loop
this page
Usability Design
primary research | user analysis
| site analysis | page analysis
| reporting
Primary research is what scientists do. Secondary research is what you do when you learn about primary research from journals and books.
I was born thirty years too early. Much of my youth was spent in libraries searching for and consuming information. The fact that it was printed on paper and stored in bound volumes made of atoms went unquestioned. The alternative of oral storage and transmission seemed downright uncivilized. The Dark Ages.
By the mid-1990's, it became crashingly obvious to me -- and to many librarians and publishers -- that most of the information humans produce, store, and share cannot economically be printed on paper any longer. To use a less threatening analogy, think of how many actors and theaters we would need to replace every TV and movie viewing experience with a live show.
Books in the sense of ink-on-paper volumes may well survive as an elite category of information storage. But books in the sense of a coherent collection of written words will thrive on digital networks. As the designers break out of flatland, the writing will become unprintable.
Learn more about researching with searching the Internet, finding information, evaluating what you find, and harvesting it in useful form.
In the Showroom section of Gizmos, Inc., learn about information itself: what it is and how it is organized on the Internet.
In the Search section of Ricci Green's Town Hall, ...
Every teacher, writer, and marketer was a student, reader, and customer first. Much of their understanding of their audience is intuitive. Gizmo-makers, however, have very little experience as gizmo-users. Especially for us old dogs, more formal audience analysis pays off in more usable gizmos.
The basics aren't hard. Without fancy equipment, you can do quite a bit on your own to make better websites.
The best resource on the Web is The Usable Web. It's just a list of categorized and lightly annotated links, but it is cared for by Keith Instone, who stakes his public reputation on the site's remaining current, interesting, and informative. Note especially the page of Issues and its section on Ecommerce.
By traditional, I mean only the past couple hundred years. The process is cyclical and iterative; progress is incremental. It is based on repeated failure and flawed successes that are continually corrected. Ideally, the process has four phases:
1) Having read about a lot of other experiments in journals, I design my own experiment to extend or confirm earlier results.
2) The design starts with a hypothesis. To test it, the design controls for variables. The process objectively measures.
3) After I logically analyze the measurements to prove or disprove the hypothesis, I write up my findings and submit the article to a peer-reviewed journal. Perhaps I read it aloud illustrated with PowerPoint slides at a conference.
4) If published in the journal, it is read by other researchers who set up their own experiments to extend or confirm my results.
This cycle is measured in years but it has produced an enormous body of valid and reliable scientific evidence. Some of it flies in the face of commonsense to say nothing of our far older and richer traditions of folklore, myth, and religion that explain many of the same things.
One of these far older traditions has also contributed immeasurably to human culture. In the tens of thousands of years before the scientific method, we used trial and error. That is, we made something and tried it out. Whether it worked or not, we watched people using it, we used it ourselves, and we asked, "What if ...?" Then we made another one, sometimes better, sometimes worse. Think of the sail, the plow, the bow, and the oven, as well as the city, the picture, and the idea.
This cycle is measured in minutes or days, as long as it takes me to build a new one and try it out. As opposed to the scientific method, which we don't learn until later in life (if at all), we learn trial and error as infants.
This research lab combines the two traditions of primary research: scientific method and trial and error. The server logs provide objectively measured evidence of your behavior. The forms linked below in the categories of:
user analysis
site analysis
page analysis
provide subjective self-reported evidence of your behavior, feelings, and opinions. In addition, I observe students, the primary audience, using Ricci Street in class and I gather anecdotal evidence from them in email, chat, and end-of-course self-assessments.
This research lab will start collecting user information with:
Basic Skills
(version for Walt Kolt's classes)
The Demographic Environment
Faculty
Needs Assessment
By loop, I mean the iterative incremental process by which we're all learning how to use the Internet. To the extent that education is a business and you are a customer, then Ricci Street is customer service and my maintaining it is customer relationship management. The more I know about you, my customers, the better I can adapt Ricci Street to your needs.
For other purposes, I may need to know about you personally and individually. Here, I need to know about you only in the aggregate. You aren't giving up much privacy that way. I hope you, in turn, respond candidly and honestly. I am listening and your response does make a difference.
Let's take my industry -- education. If I were to visit a series of college web sites, I'd find evidence that they are all grappling with the Internet.
At one end, the PR department has clearly felt the need to keep up. The site is brochureware. The PR folks have taken a print brochure and used the same text and graphics and, as far as they could, layout. They probably measure success by the degree to which the web site looks and behaves like the brochures they're used to. Evidence would be in the metatags, the contact person, the slow-loading graphics, the file names. It would also be in the lack of off-site links. They're on the Web but they don't quite get it.
At the other end, the college web site is a portal. Students can
register for classes
order pizza
shop at the
school store for everything from textbooks to t-shirts
download
course materials
browse the
library's catalogue and databases
send email to
professors
learn about
professors' professional activities
read the
college "newspaper"
find
up-to-the-minute college sports scores and schedules
find
up-to-the-minute social schedules
learn about a
class they want to take next semester
submit
homework
take tests
discuss campus
and academic issues in threaded forums like the Bistro
maintain a
personal web site (www.college.edu/students/yourname/)
attend
meetings of student committees
attend
meetings of course project teams
share
documents, including audio and video, with classmates on the campus network
share
documents, including audio and video, with others on world-wide peer-to-peer
networks
check class
schedules and transcripts
pay tuition,
fees, and parking tickets
get help with
their computer problems
chat with
friends
Students can even find the brochures that the PR office put online. If the students live off-campus, they can dial in, so the college functions as their ISP.
If you were a parent of a high-school junior or if you were an employer, that is, a potential customer of the college, where would you want your child or fast-track employee to go to college? Now look at the industry you've chosen for your BigBiz project.
If you were a potential customer or supplier or business partner, what would this site tell you about how well the company is grappling with the Internet?
While you're exploring and discovering, you also need to be training your eye. I'd like you to look at webs on two levels: content and technique. Take, for example, a site like our course web. Of course, you look at it as a student getting information from the instructor. I'd also like you to put on your new digital developer's hat, open the course web in FrontPage, and look at shapes and colors and images ... and at HTML code.
Other than report summary site stats at Ricci Green, I don't do any formal analysis of Ricci Street. Nor do I read papers at conferences or publish in peer-reviewed journals. Why?
Professional conferences are designed to disseminate information (among other things). I'm doing that at Ricci Street and gaining a wider and more targeted audience than I ever could at a conference.
Journals are too slow. By the time it makes its way to a library shelf and gets picked up by another researcher, it's so dated that it's useless. Take a look at First Monday, a highly respected online-only journal. Its publication cycle is more like that of a newspaper than an academic journal or even a popular magazine. Recent First Monday articles have an editorial history at the end. They are often written only weeks before they are available on the Web.
In First Monday's archives, look at some of the articles from two and three years ago, the ones that would about now be appearing in print journals in the library according to the old system. Some of those early First Monday articles are curious, unintentionally funny, and downright misleading if you don't keep in mind how much the Internet and Web have changed in the last few years.
What I do is plow my research back into Ricci Street. The cycle time is quick, often overnight. The process is closer to trial and error than to formal experiment. However, I have a ton of objective evidence in the server logs and a nifty piece of software by WebTrends that lets me slice and dice it every which way. I also have the results of the user, site, and page analysis forms in this section. They tell me more than I can handle.
While I'm saving most of this primary research, I'm still discovering how to best use it. Suggestions are always welcome.
===
Business 2.0's Website Usability
Keith Instone's UsableWeb
National Cancer Institute's Usability.gov
|
||||||||||||||||||||||||||||||||||||
