We've all found ourselves on a Web site that's the online equivalent to a desert island. Either the confusing site navigation makes us feel shipwrecked, or the constant assault of pop–up windows and flashing images make us feel as though we're getting pelted by rain. Though Web sites that are difficult to use may have any number of underlying design or content flaws, the outcome is always the same: poor site usability.
What is "usability"?
Usability is not something you attach at the end the development line, like another coat of paint or a racing stripe. Usability is something you design into your site. It should be a documented, measurable requirement, just like efforts to ensure that your site is compatible with different browsers. Since each Web site is different, however, and since Web site owners have divergent ideas of what they want their site to achieve (a place to make online sales, to make offline contacts, to find prospects, or to distribute information to a group of experts or fellow hobbyists), the particular attributes that make sites usable vary.
Nevertheless, all good Web sites have certain things in common. By incorporating the following tips and conducting basic testing, you'll make dramatic improvements in your site's usability. It's worth the time and, if necessary, the investment. For every dollar spent on usability, after all, a business stands to realize a benefit of between $10 and $100.1
6 elements to Web sites that "work"
Every Web site that delivers a clear, pleasant user experience has the following characteristics.
- Visiblity of the core process — Great sites keep things simple. If you're selling yo–yos, then the user should know they're on a site about yo–yo's, and they should be able to start browsing for yo–yos and even start buying them from the homepage.
- Clarity of links — Each link on a Web site should clearly explain where it is going and what it's going to deliver. If you want to provide the user with contact information, the link should say 'contact information', not 'more information', or 'Let's talk'.
- Clarity and timeliness of instructions — Instructions should be just–in–time; that is, they should appear where and when the user needs them. This is especially important in sites that offer users a lot of options reflected in complex task processes. Make sure that instructions are placed next to the functions or controls they refer to and are brief, clear, and free of spelling errors.
- Error prevention — Anticipate potential errors and include instructions at those points to preempt mistakes or wrong turns.
- Error recognition, diagnosis, and recovery — When errors do occur, it should be made clear to the user what needs to be done to fix things and then proceed. Use clear text to explain what's happened, and let the user easily return to where they were before the error occurred. If they've failed to enter one bit of information, don't make them re–enter all of their information.
- Consistency — Good Web sites use the same controls, navigation, and features from page to page. Continuity makes it clear to the user that they're on the same site and makes it easier for them to find the same core functions wherever they go on the site.
For every dollar spent on usability, after all, a business stands to realize a benefit of between $10 and $100.1
How to conduct a usability test
Once you've gotten the above elements in place, you're ready to initiate usability testing. Before beginning, it's important to remember that there are three elements against which usability is gauged.
- Effectiveness — This describes how well, or accurately, users are able to accomplish the goals they set out to achieve. For example, if users go to a Web site that sells kites, that site would be considered effective if each of those users were able to find and purchase, with a minimum of error and confusion, the specific kite they wanted.
- Efficiency — In the context of Web site usability, efficiency describes the speed with which users can complete their tasks. To continue with the above example, if users are able to find and purchase a kite quickly and easily without errors, redundant clicks, or dead ends, the site would be considered efficient.
- Satisfaction — Satisfaction is a measure of users' feelings during a transaction — their level of confidence as they complete a task, and their level of confusion or frustration with site instructions and transaction processes. In the end, will users be happy to use this site again, or will they avoid it?
To test your site's usability, use the above three benchmarks as your guides. Then follow these three steps to develop a usability test:
Step 1: Identify the type of people you expect to be your primary users. To accurately test your site, you need to identify who your customers are and gather a small, representative group to test your Web site. Is your customer group familiar with computers and the Internet? Will your customers be familiar with the products or services available on your site, or will they be novices? Are they likely to fall into a particular age group, have a specific level of education, work in a specific job or industry?
Admittedly, locating ideal test subjects may not be easy. You might have to entice them with free food, money, gift certificates, or t–shirts.
Step 2: Identify the primary tasks users will be performing on your site. These tasks will be at the heart of what your Web site is about. The key tasks users perform on a typical e–commerce Web site are:
- Searching for a product
- Placing products in a shopping cart
- Paying for products.
Step 3: Set the performance conditions. This relates to steps one and two. What kinds of scenarios and conditions do you expect your users to be facing when using your site? Will they be in noisy, frenetic offices or airports, or will they be working from the privacy of their home? Will users typically want to browse around your site, or will they want to complete a transaction and get out? Once you've identified the conditions under which most people will interact with your site, you should try and do your best to reproduce those conditions in a test setting.
Resources, next steps
Usability testing is not rocket science, but it does require preparation and planning. Be sure to take advantage of the following resources to help you plan and conduct initial and ongoing usability testing:
The Usability Professionals Association. The resources section of this site contains links to usability testing methods, user interface design style guides, Web interface design style guides, and much more.
The Society for Technical Communication, Usability Specicial Interest Group. This site contains bibliographies, Web references, downloadable files, and other usability–related materials.
Usableweb.com. Contains links to a wide range of usability resources, as well as a Top Ten of usability books.
Amazon.com. Enter 'usability' in the Amazon.com search engine and you'll discover over 50 publications on the topic, many of which contain the latest self–help and do–it–yourself methods.
Consultants: There is an increasing number of usability consultants promoting their services, and, like all consultants, some are better than others. I suggest you contact the Usability Professionals Association for a listing of UPA consultants in your area.
About the author: Tom McCann is an Interactive Usability Manager for Staples.com.
