Re: "Good" Web Pages

["Jared M. Spool" <jspool -at- UIE -dot- COM>]

On 29 Jan 97, mmarkley -at- MICRON -dot- COM wrote:

> What is "good" when it comes to Web pages/documentation?

We are in the process of conducting a lengthy study on the usability
of websites.  We've studied 9 sites, including C|net, Disney,
Fidelity, HP, and the Olympics sites.  We took the data from dozens
of hours of user observations, compared them to a detailed analysis
of the composition of each site, and looked for any correlations that
would tell us what makes one site more usable than another.

The initial report (due mid March) will be over 100 pages --
obviously I can't describe all the results here.  But here are some
quick bites:

C|net, which someone touted as a good example on this list, came
second to last in our study.  (Disney was last, HP & Olympics were
first & second).  There are many reasons as to why this is, but the
key ones were that the home pages offered very little direction on
content.

Edmunds came in third.  What's interesting is that Edmunds has
virtually no graphics -- it's all text.  It beat C|net, Disney,
Fidelity, Inc and Travelocity -- all very graphic intensive sites.

One of the measures we did of the site was how "readable" it was.  We
took the text and ran it through readability measures, such as
Gunning-Fog & Flesch.  We found a surprising result:  the less
readable the page, the more likely people would succeed at finding
the information they needed.

Another measure was the amount of whitespace.  We found a similar
result:  the more whitespace, the less success people would have
with finding the information they needed.

Our theory is that people don't read.  Instead they are skimming
(which someone else on this list recently aluded to).  Readability
and whitespace doesn't contribute to skimming.

Other findings:

Download time is not the issue everyone thinks it is.  People are
happy to click on pages while they are still downloading.  Just don't
put any content in the graphics.

Graphic design had very little effect, positive or negative on
overall site usability.

Navigation that wasn't tightly linked to the content was less usable
that tightly linked navigation.

The best site (HP) scored only 159 out of 343.  While it was almost
three times better than the worst site (Disney, score 63), it still
wasn't even half of the best possible score.  Our take: we have a
long way to go to get to an optimum design for usability.

If you are interested in this topic, I'll be giving a presentation at
the Winhelp conference this week in Seattle on this topic.  I'm also
presenting it at the Web Developer's conference in San Jose at the
end of the month.

Also, we'll be posting details, in small chunks, on our free email list:
UIETips.  (You can subscribe by sending a message to
UIETips-Request -at- uie -dot- com with the word SUBSCRIBE in the body of the
message.)  Plus, a detailed report will be available in mid-March and
we'll publish some of the results in Eye For Design, our newsletter.

Hope this little bit of information was helpful.

Jared
==========================================================
 Jared M. Spool                User Interface Engineering
 mailto:jspool -at- uie -dot- com     800 Turnpike Street, Suite 101
 (508) 975-4343                   North Andover, MA 01845
 fax: (508) 975-5353                                  USA

      If you send me your postal address, you'll get
     the next issue of our newsletter, Eye For Design.
==========================================================

 TECHWR-L (Technical Communication) List Information: To send a message
to 2500+ readers, e-mail to TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU -dot-  Send commands
   to LISTSERV -at- LISTSERV -dot- OKSTATE -dot- EDU (e.g. HELP or SIGNOFF TECHWR-L).
  Search the archives at http://www.documentation.com/  or search and
browse the archives at http://listserv.okstate.edu/archives/techwr-l.html