PRIVACY STATEMENT |  ABOUT US | HELP | LOGOUT

• Advertising • Animated GIFs • CGI Programming • Consulting • Credit Cards • E-commerce • Graphic Design • Hosting • JAVA/Javascript • Site Statistics • Shopping Carts • Shipping • Tools &     Software • Style Guides • Training Get Free $$$ Get paid for the cool content on your Web site!.   WEB SEARCH:

Web Design Considerations Visitors won't suffer a poorly designed site. The experts at the Neilsen Norman Group show you how you can improve your site's usability.  Usability is often the most neglected aspect of Web sites, yet in many respects it is the most important. If visitors can't use your site, they will leave and never become customers. The Web gives people too much freedom and too many choices; no one will suffer a poorly designed site. To make your site usable, you  need to involve potential customers in its design.  The pressure to get your site up and running can make it seem impossible to add any steps to the development cycle. But can you afford to launch a site that bombs?  When people leave bad Web sites, they rarely come back to see if you got it right in Version 2.  Ideally, you should involve users throughout the design phase. Life isn't always ideal, but you're still better off getting some feedback before you launch. In the following pages, we'll first show you how Ideas.com reaped big benefits from a quick round of prelaunch usability testing and then give you tips on how to get  usability feedback on your site quickly.  Ideas.com, a "marketplace for ideas," helps match people's ideas with companies that need them. (Polaroid's sticker-printing camera idea, for example, came from an Ideas.com user.) A couple of weeks before Ideas.com's initial launch, we did a usability study for the site. We had five users go through typical tasks and report on how easy or difficult they were.  This relatively small time and resource investment had big payoffs. As the examples on the following pages show, we vastly improved usability in some areas with some quick redesign work. The fixes were simple, but  had they not been implemented, many potential customers might have left the site. Several of the problems surfaced on common Web site drop-off zones, such as the home page and the registration area, where the danger of losing visitors is greatest.  Quick ways to improve your site's  usability You can effectively collect usability feedback and improve your site, even if you don't have resources, a lab, or money to hire consultants. Invite five people to your office (at different times) and ask them to use your site -- regardless of what stage it's in. Watch them  as they surf.  Don't wait until your site is completed before testing.  Test now, whether your site is live or just a paper site map and some pages. The earlier you collect and respond to user feedback, the better your site will be.  There are a few things to do when conducting even the most basic studies. The following steps will lead you through the process.    1. Determine goals and write the user tasks Decide what you want to learn. See if people can order a toy, for example. Based on your goals, write tasks for the user to perform. Give your testers enough information to proceed without actually revealing anything or documenting the site. Give them realistic tasks, such as: Send your 10-year-old nephew a birthday present. Don't use specific site terminology such as buy, add to shopping cart, check out, store, or search results.  Instead, let the users decide what products they'll buy  and how to go about buying them.  Number and write each task on a separate page. This helps you track what the user is trying to do. Also, this lets you present a new page each time the user moves to the next task, allowing some sense of accomplishment even if the person could not complete  the task.    2. Determine the user profile and schedule the sessions Choose participants who represent the target  market. Testers should have experience levels similar to those who will use the live site. Decide what traits are important, considering Web experience, operating system knowledge, occupation, experience using competitive sites, and familiarity with your site.  Typically, you need only four to six people to test a site. Do not bring all the users in at the same time.  Spend between 20 minutes and 2 hours with each tester, depending on how much you want to learn.  Confirm the testing time with all your testers, and follow up with phone calls the day before the session.  Be prepared to offer some incentive. Popular honor ariums include cash or checks, sweatshirts, software, or pens with product logos.    3. Conduct the sessions You can run the tests any place quiet enough to hear what the user is saying.  If you don't have an observation room (typically a room with one-way glass so you can watch the testers without disturbing them), limit the number of observers to two or three and tell them they need to be silent.  Before the session, set testers' expectations and make them feel comfortable. Explain that you want them to think out loud, telling you what they are trying to accomplish and what they do and don't understand.  Ask them to read each task out loud before attempting  it. Make clear that they are not being tested, they are testing the Web site so you can learn how to make it better. Tell testers you will be taking notes and will try not to talk with them during the sessions.  Resist the temptation to talk or help the users.    4. Evaluate the data Pay close attention to areas where users were frustrated, took a long time, or couldn't complete tasks.  Respect the data and users'  responses, and don't make excuses for designs that failed. Also, note designs that worked and make sure they're incorporated in the final product.  Implement changes and repeat the steps above as long as the site exists.   5. There's a lot more you can do Although quick studies will help you, there's a lot more you can and should do to improve your site's usability. Methods will vary based on the data you collect and the resources at your disposal. See Jakob Nielsen's book Usability  Engineering for detailed coverage of user observation methods.  For interaction design, the most important thing to remember is that watching people work is much more telling and gives truer information than focus groups or surveys. In the latter methods, users are asked to do two almost impossible things: realize when they encounter a usability problem, and then remember and  report it. Focus groups add another element of  complexity, because the participants influence one another. For these and various other reasons, you're not likely to get completely accurate reports.  Surveys and focus groups have benefits, but are typically less  informative and can be misleading when used to influence design.  Many site operators still use launch as their first round of user testing. Unfortunately, undoing a bad launch is just about impossible, and even small usability problems can drive visitors away from your site. The earlier you  integrate user feedback into your development  process, the better your chances for success. Even a small number of users, like the five who gave us feedback on Ideas.com, can make a big difference in a site's success.

Already a customer? Sign in to see what's New for You . Tree of Life.Org MyOwnBizKit.Com More New Releases Coca Cola Bottling Spanish Conversion Web development project BMG Music, Inc. Spanish Conversion Web development project See all the Movers & Shakers View customer's Wish Lists.    Special Features • Buy or redeem a gift certificate • GMCD Electronic Publishing, Inc. Anywhere • Get e-mail recommendations from our developers.

Where's My Stuff? • Track your recent order. • View or change your orders in Your Account.

Need Help? • Forgot your password? Click here. • Redeem or buy a gift certificate. • Visit our Help department.

Pricing  |  Sites  |  Software  |  Shipping  |  Billing  |  Contact Us  |  About Us  |  Help Site Map  |  Job Opportunities  | Privacy Statement Copyright © 2000, 2001 GMCD Electronic Publishing, Inc.  All rights reserved