It seems that in any new endeavor--and for most companies, e-commerce is as new as it gets--we have to go back
and reinvent almost every quality concept and method. Twenty years ago, when total quality management first reached service companies, there were numerous authors and consultants who claimed that
service quality was totally different from manufacturing quality. New terms were invented, new methods created and many special workshops developed to teach this "new" area of quality
Now we wonder what all the fuss was about. When we really begin to under-stand the new area, we find that the same concepts, methods and tools apply. This is
definitely true in designing a good Web site.
The first idea that must be established is why we are designing a Web site. What are our objectives? What value are we trying to
add to our company or organization? Are we trying to increase sales, attract more customers, provide information to existing customers, entice more people to our other facilities, sell products
directly, reduce costs of business transactions or provide information to employees?
The first step after clearly stating our objectives is defining exactly who the customers
are for this Web site. Surprisingly few business owners have thought this through; too often they try to be all things to all people. Only when we have stratified the potential users of the site
into well-defined, homogeneous groups should we even attempt to start designing the site. Different groups will usually have widely varying needs.
The second step is defining
the needs of each of these groups. What information are they seeking? How are they likely to search for this information? What type of access will they have (high-speed wideband or clunky old
modem)? How sophisticated are their searches? How informed are they about the subject? What jargon is acceptable, and how much detail is needed or will be tolerated?
we have solid, reasonably complete answers to these questions should we attempt to start designing the Web site. At this stage, we'll need a variety of talents on the design team. The Web site
structure or information architecture is the biggest issue. How will the various data bases and access pages be organized? What will appear on the home page and the second level pages? How can we
lead users in three or four mouse clicks to the exact information they need? Research shows that we need to keep the search to four or fewer steps with no more than 16 seconds per page. The pages
need to be uncluttered so the users can quickly find what they need, but pages should also be information-rich enough to minimize the number of steps.
A key member of the
design team at this stage is the graphics arts specialist. How the site looks is critical--but only if the architectural structure is right and the needed information is present. Pretty sites
that don't work are disasters. Another key team member is the information specialist, the person with detailed knowledge of exactly what information we're providing and how it will be used. For
many organizations this will need to be several people, perhaps a large well-organized team of creators and reviewers.
This team now creates the process features, defining
carefully the information that will be provided to each customer segment and how the customers will access the information. This third step is critical; it's the equivalent of product design.
These features are what we're actually "selling" to our customers. At this step we need intensive interactions with focus groups or samples of intended customers to ensure that we're
designing a product that will meet their needs. Too many organizations fail to do this well. One organization receiving more than 900,000 hits per day on its site was relying on a survey of less
than 500 customers to understand needs, use patterns, common behavior and satisfaction.
The fourth step is the software programming and hardware selection for the actual
functioning of the site. Unfortunately, many organizations begin here, skipping steps one to three. They rush into getting a site up and working without thinking about what it is they'll be
offering. A critical issue at this stage is designing for maintainability; that is, making it easy to manage the site; update it; and respond to questions, orders, customer complaints and other
The fifth step is implementation and quality control. What measures will we use to monitor how well the site is meeting the needs of our customers and our own needs?
Are we going to count numbers of hits? Will we measure orders received? Will we measure the time it takes users to find the needed information? Will we measure "stickiness"--how long we
can keep customers at our site? (For some organizations, this is a desired positive measure, for others it is a negative measure.) Are we able to measure blockages, dead ends and failed links?
Are we able to make competitive measures against similar sites? How will we measure freshness? And most importantly, did our site meet our original objectives? Did it add the value we thought we
Designing and developing a good Web site is fiendishly difficult--just like designing and developing a good product. But whoever said quality was easy?
About the author
A. Blanton Godfrey is dean and Joseph D. Moore Professor at College of
Textiles, North Carolina State University. Godfrey can be contacted by e-mail at firstname.lastname@example.org .