Developer Center Article

Sites that are easy to use, software that supports user goals, and loyal customers who renew subscriptions and refer their friends are all signs of a successful design implementation. Every production team should strive to achieve such results.

User-centered design activities—such as field studies, prototype tests, and customer modeling—require pre-production costs. While these tasks can add to your initial production overhead, usability testing will greatly improve your deliverables in the long run. Don't skip this important part of the design process. Even if you cannot afford expensive laboratory rentals and materials, it is well worth the effort to invest some resources into low-budget methodologies. The benefits of evaluating your plans and revising them based on user feedback should not be underestimated.

All activities inherent in user-centered design are learnable, and team participation is critical. You must be realistic about what tasks need to be outsourced in terms of research planning, execution, and analysis. The best consultants will teach you skills that you can do yourself the next time.

Building it right the first time will reduce your total development budget. You can choose to focus on what can be built and what looks cool„or you can focus your development efforts on what your users really want.

It is important to ask yourself: "Does this site or service solve a problem that users currently have?"

Visit users in their everyday environments. Interview potential customers in their offices, homes, schools, leisure spots and shops. It's not enough to listen to what people say they do, as in focus groups. By going into users' environments, you can observe what they actually do. Your mission is to discover patterns of usage.

Traditional usability testing teams ask the question, "Can users accomplish directed tasks?" By contrast, successful implementation testing teams should ask the question: "Would users choose to adopt this service?"

More than ever, new interactive services must persuade users to assume new behaviors in a world of online and offline competitors. Anthropologists increasingly provide technology and product teams with knowledge about customer motivation and behavior. Qualitative research must balance the desire to test specific hypotheses with an openness to discovering the unexpected.

Customer visits provide insight into the technical, emotional and lifestyle preferences of your users. Use digital video or still photography to record user interactions and environments (see Figure 1). This rich information drives strategic and conceptual decisions about your site. Will your service work on your usersÍ platform„and take account of their connection speeds, plug-ins, and tolerance for learning new applications? Will it fit into their lives? Is your service easier and more valuable than online or offline competitors? Will users intuitively adopt new behaviors? Will your services be integrated with other accepted activities?

Whom should you visit? Certainly not your client or managerÍs spouse, or anyone who has a vested interest in your projects' success. Define the typical user, and be sure to exclude power users„unless they comprise your main user group. Six users are usually enough to establish patterns of behavior and motivation. Recruit interviewees through word of mouth and community sites, or hire a professional recruiter. You may also need to outsource the creation of a screener, an observation guide, a lead researcher and analyst, or report writing.

Make customer models

When creating a model of your anticipated user, use the feedback gathered from customer visits as well as your internal customer knowledge. Include log analysis, sales data, and prior quantitative and qualitative research to build user profiles and scenarios. In choosing a target audience, cast as wide a net as possible to increase your potential audience. Go beyond early adopters and reach for a mass audience.

Create scenarios and profiles that focus on user goals and constraints. Whom are you designing for? What are their goals and aspirations? What elements of your site will be most relevant for each user type? See Figure 2 for an example of relational user profiles that show radically different users of a mock Internet children's toy.

The motto "know your user" applies equally to customer modeling as it does to the development process. Choose a format that makes the most impact in your work environment. Customer models range from simple posters and task flows to mental models and detailed scenarios.

Create early prototypes

Perhaps you have a few ideas about how to design your site. You must factor in technical feasibility and business viability. But don't forget to consider what will make your site or service most compelling to your users.

DonÍt guess what will be successful. Prototype early and often in order to allow your users to help you refine your strategy. Focus on their goals, not on system requirements. Get early feedback before investing time and money into development.

Prototypes can be hand drawn and shown to six users in a single day. You can quickly mock up alternate home pages or study competitors' services. Invite research participants to tell stories about comparable offline or online services, explore your prototype, provide feedback, and even redraw your paper screens.

Iterate prototypes with your users

A round of user testing can provide a wealth of new ideas for improving your service and interface design. Spend the time necessary to incorporate this knowledge into improved prototypes.

Test and retest. Repeat these steps until you are confident of achieving business goals and satisfying real user needs. Design iterations can progress from a hand-drawn paper prototype to a wire frame to a final visual design (see Figure 3). This process provides an opportunity to hone interaction, nomenclature, and experience. There's no need to launch a site without knowing how your users will react. There is no need to launch a site without knowing how your users will react.

Build on your successes

Sometimes you have to jump in feet first in order to begin the user-centered design process. You may have to start in the middle, or apply last-minute test results to future releases. Changing design and development processes is difficult.

Whether you start with a field study or prototype evaluation, invite all stakeholders to participate as note-takers, observers, and co-interviewers. Communicate your findings to everyone and demonstrate how user knowledge has improved your UI and business strategy.

Finally, connect user-centered design changes to clear business metrics and site goals. Include variables such as the number of visitors, length of time on site, product sales, subscription renewals, and peer-to-peer referrals. Discover what works and let your audience be your guide.

References

Braiterman, Jared and Richard Anderson, 2001. Strategies to Make E-Business More Customer-Centered in The Business of Usability (Springer-Verlag, London).

Braiterman, Jared, Sasha Verhage, and Randall Choo, 2000. Designing with Users in Internet Time in Interactions, September-October, pp. 27-34.

Cooper, Alan, 1995. About Face: The Essentials of User Interface Design (Hungry Minds, New York).

Wixon, Dennis and Judith Ramey (eds), 1996. Field Methods Casebook for Software Design (John Wiley & Sons, New York).

Kuniavsky, Mike (Forthcoming). Practical User Testing for the World Wide Web.

Anthropology Resources on the Internet

Human Computer Interaction Bibliography Project

The HCI Index

Usable Web

About the author
Jared Braiterman, Ph.D., guides executives, investors, and product teams in learning directly from customers and using that knowledge for product innovation and business strategy. He received his degree in Anthropology from Stanford University in 1996. Clients include Adobe, Leap Frog, Shutterfly, Listen.com, Hewlett-Packard, eXpn, Electronic Arts, and Citibank. More information can be found at jaredresearch.com.