CHI-WEB Archives

ACM SIGCHI WWW Human Factors (Open Discussion)


Options: Use Classic View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Topic: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Louise Penberthy <[log in to unmask]>
Thu, 13 Feb 1997 10:45:42 -0500
text/plain (73 lines)
At 10:15 AM 2/13/97 -0500, Levi_M wrote:
>A colleague and I have been having a sporadic disagreement concerning
>whether a keyword search of our Web site indicates a failure in the
>site's structure.  I would be interested in your insights.
>The context is that our Web logs track all user invocations of the
>"search our site" feature, including the keywork(s) entered in the Web
>form.  I believe that we can and should interpret multiple users
>searching for the same or similar concepts as a failure in site
>design, and use this evidence as one input to a redesign effort.
>Most intra-site search engines -- ours, sadly, included -- are so
>hideous from a usability perspective (unclarity of matching algorithm,
>too high or too low volume of responses, inadequacy of responses) that
>only a truly desperate person would turn to them.  If the site
>hierarchy made an efficient navigation path possible, use of keyword
>searches would be unnecessary, and, correspondingly, would dwindle.

Another issue is that the menu items designers use on their Web sites are
not always intuitive (even assuming an intuitive menu item can be found for
every screen that can be navigated to).  And the menu items are not
consistent from one site to another, largely due to the fact that the Web is
still relatively young and not stable from a design point of view.  As part
of my company's efforts to start redesigning our site, I recently did a lot
of surfing of similar companies' sites.  The variance is astonishing --
which is evidence of creativity, but also of problems.

Part of the reason that people use search engines may be that if they come
-- for example -- to a company's site, and want to know whether they have
any job openings, it's not obvious where to go to find the listings.  One
solution is to have, on the home page, a way for a user to see a description
of what can be found at each page of the site.  A list, for example, or Java
applets that provide feedback on mouse rollover.

Other pages on the site would, of course, have to have an obvious link to
the home page.

>From looking at the logs, I see two main patterns.  Some users
>immediately invoke the search engine, without exploring the link
>hierarchy.  I attribute this to learned behavior from other poorly
>designed sites, and do not feel that this necessarily reflects poorly
>on us.

Having not seen your site, I can't say for sure.  But see my comment above.
A site doesn't have to be poorly designed to make people want to use a
search engine.

>Other users click around for a while before starting a search.  This
>is where I feel we can draw some conclusions.
>My colleague disagrees.  He believes that the search mechanism is an
>independently valid and appropriate navigation strategy, and its use
>should be seen as such.

I agree with you.  If users are looking around, and then turning to the
search, they may have been unable to find what they wanted.

OTOH, it may be that they searched around to get a feel for your site and
what they could find.  Once they established that, they felt confident
enough to start a search.  That might suggest redesigning the pages on your
site so each gives a better feel for the site as a whole.  Do people tend to
start at your home page, or some other page?

-- Louise

Louise Penberthy                                Voice:  (404) 875-5620
Information Designer                            Fax:  (404) 870-0895
Atlantis New Media, Inc.                        [log in to unmask]
390 Trabert Ave., Atlanta, GA  30309