When we create customer-effective Web sites, we are concerned with developing Web sites that make it easy for customers to do what they need to. What customers want to do is the context for development and must also be the context for testing.

To find out what customers want to do, and how they actually need to go about doing it, we must find out about customer behavior. We need to understand the big picture as well as the specifics of what customers will want to do on a Web site. And that requires some smart planning and execution across the organization and the Web site development company.

Customer needs drive the development process and are our touchstone for success throughout that process. We need to talk to customers, up-front, before development and then pull them back in at critical points in the development process prior to launching a Web site to the public. The testing gets more and more specific as we go. A lot of organizations run into problems when they try to test generalities when they need specifics, and vice versa. General customer feedback is appropriate early in the development process, but later on when development has progressed to specific designs and functionality, specific customer feedback will be required. The level of the testing, its timing, and the techniques used all affect our ability to get the right information when we need it.

If we don't know what we need to know about customers, we will not develop customer-effective Web sites. We may develop good Web sites, but if customers can't, or don't, adopt them into their service relationships, then we've failed.

So, we test our ideas with customers in an attempt to offer them the best e-services we can. And, this implies an iterative process where we have to potentially redevelop concepts or parts of a Web site and then retest them.

Iterative processes often send organizations and Web site developers into a spin. It's not surprising when we are all working with tight deadlines. We simply don't have the time to redo, redo, redo. Do we?

As with all things, it's a matter of striking a balance. It is possible to work against tight deadlines and still benefit from customer testing. That said, if you have no intention of revising what you're doing as a result of customer feedback, it's probably not worth asking customers what they think in the first place.

Of course, we don't have to deliver a perfect e-solution from day one. Customers will learn with us, and help us along the way, as long as we give them something useful in the first iteration of our Web site. And therein lies the problem. Too many organizations continually save revisions to "the next release" and launch a Web site that offers little or no additional value to their customers. Customers don't think in release; they think about what they need to know and do, now, not later. If you provide customers with useful content and functionality in the first iteration of your Web site, and then continue to deliver additional useful content and functionality when it is needed by e-customers, you can keep them on board.

An organization needs to take e-customers' priorities into account when deciding what content and functionality will be released and when. If e-customers are all asking for a specific service function, such as view and pay their bill online, because it is the most important thing they want to be able to do on an organization's Web site, then this should be the first thing to be developed by the business.

Sometimes e-customers want things businesses can't deliver straight away, and this needs to be understood and managed by the business. Sometimes e-customers want things that can be delivered straight away, but they would have been over-looked had e-customers not been consulted. Testing, therefore, is necessary to really understand e-customer priorities and how businesses can create value for e-customers from day one.

If we set the right direction at the start, then we avoid having to reinvent during the development process. Reinvention, late in the development process, is a waste of everybody's time and resources. More often than not, organizations will roll out an inadequate Web site, rather than redo what has already required a lot of time and money. And that's a risky move, if you know you're not on the right track with your customers. We may lose our customers by providing our competitors with the opportunity to offer them something better.

Customer testing should lead to invention, not reinvention. It's just a matter of good management and timing.

And, of course, customer testing can also be a way of fixing obvious usability problems we miss. Many of the usability mistakes we make relate to the way information and functionality is presented on Web sites and can easily be fixed. Being aware of the 17 customer directives covered in this book will give businesses and Web site developers a head start on avoiding some of the most common, and perilous, usability mistakes.

Jodie Dalgleish
Vice President
Jodie Dalgleish Vice President for Gartner's Consulting Services. As Vice President and an e-business specialist, she draws on 10 years of senior marketing management and professional communications consulting experience. Ms. Dalgleish has three years of specific, hands-on e-business expertise, which she gained after setting up her own Web consultancy business to meet the demand for dedicated, end-to-end, strategic e-business consulting services. Ms. Dalgleish has a master's degree in business and administration from Victoria University in Wellington, New Zealand, and is author of the highly acclaimed book entitled "Customer-Effective Web Sites".