Usability Testing

Nicholas Papps
MCCNM 336
Fall 2003

1. Introduction
2. Defining Usability Testing
3. The Importance of Usability Testing
4. Preparing and Conducting Usability Tests
5. Characteristics: Good sites/Bad sites
6. Conclusion
7. References

Introduction

Before we can define what usability testing is, it is essential that the understanding of usability come first. Usability means that the people who use a product can do so quickly and easily to accomplish their own tasks. While keeping this in mind there are a few other things that can help make this definition more clear. Usability above all means focusing your ideas of development on the users. A product like a web site has to be easy to use and allow users to accomplish tasks. Professional web designers have to know and understand the people or potential people that are going to use the product. There is nothing that can reveal more information about the usability of a site than the users themselves (Dumas, 1993, p. 4). When developing a website an important goal for a designer is to make the website easy to use. This is always determined by the users of the site in terms of the time it takes to accomplish their task, the number of steps it take to get there and the success they have in anticipating the right action to take. Still, the main focus in web design is to recognize the users goals and develop a product that makes it easy for them to complete (Dumas, 1993, p. 4). Although usability and usability testing covers just about all areas in the creation of products, this paper is going to focus primarily on website design.

Defining Usability Testing

Usability testing is improving the usability of a website that is being tested. The definition of usability testing is also the primary goal. Another important goal for site designers is improve the process by which products are designed and developed in order to avoid having the same problems again in other products (Dumas, 1993, p. 22). In a broad scope, this is best completed by thoroughly testing the site on potential users. Sometimes people get the functions of usability testing confused with quality assurance or function tests which has a goal of assessing whether the product works according to its specifications. The difference here is that a new website being tested may meet all of goals in terms of specifications but may fail when testing its levels of usability (Cre8pc, 2003)

The Importance of Usability Testing

With over half a billion websites on the Internet right now; users have more choices than ever. Users do not want to be wasting their time on anything that is confusing, slow, or that does not satisfy their needs. The result of this overwhelming amount of choices and the ease of going elsewhere has caused web users to become increasingly impatient and their need for instant gratification has overall increased. If users cannot find out how to use a website in a minute or so they conclude that it won't be worth their time and they leave (Nielson, 2000, p. 10). This is where usability testing really comes in handy. It's dangerous to assume that you know what users want. The argument, "I'm an expert user, so if the system is designed for me it will work for everyone," is a telltale sign that you've introduced unnecessary risk (Maya, 2003).

Important Roles

In recent years, usability testing has taken on a much more important role than it has in the past. In traditional software development, customers didn't get a chance to test the usability of a product until it was already bought and paid for. Now there are free demos that canvas the glob. Website development has an opposite situation. Web users experience usability first and pay later in most cases. Web sites allow user to experience the site before they have committed to using it and before they have spent any money on potential purchases (Nielson, 2000, p. 10). On the web, with all the millions of sites out there, a designer is only in competition for the users' time and attention (Nielson, 2000, p. 10).

Preparing and Conducting Usability Tests

Planning is critical to a successful test and is necessary to insure a smooth and useful test. If you don't plan properly bad things can occur that a tester might not be prepared for. The first thing to ask when planning a proper website usability test is, “how long is the usability test is going to take?” This is an important question to ask because the amount of time needed has implications on budget and schedules. The main goal when planning a test is to make the actual usability test as short as possible and still be able to yield useful information from the test (Dumas, 1993, p. 22).

Length of Tests

Usability tests can take anywhere from 12 weeks to a day and a half depending on how much time is needed to gain useful information about the site. Longer testing times are usually for large companies like Microsoft or Sun Microsystems for example. Companies that follow a formal test process usually allot 8 weeks or more for a usability test (Dumas, 1993, p. 101). Having more people to perform a usability test on generally means longer the test will take to complete. For larger companies testing specialists prepare formal reports on the usability tests covering the results and projections for the site to be given to managers. Organizing data like this can easily add more time to the overall test. Usually the smaller and less formal the tests are the shorter it takes to complete test on a site. Tests that last a week or so are usually for smaller organizations to get a rough estimate of how well a sight is going to do. Shortening the time of a usability test depends on a few important factors. If there were any problems the site the developing team came across they can concentrate on that during the test. Another strong attribute to shortening testing time is to have everybody that is involved in the testing fully understands what he or she are doing and nobody is on a learning curve. Not doing a formal report on the test in general saves the most time.

Finding Users

Finding users to participate in the test is the next step. Testers will develop a profile of the actual and potential users of the site long before the test actually takes place. Understanding potential user characteristics like demographics is important. Factors like income level, education, and age. Most sites are designed for users of average mental and physical abilities. But, there are some sites that cater to people with special characteristics such as children or people with disabilities. Also, the amount of computer experience each user has should be understood. The selection of people to use for the usability test should reflect the potential users of the sight to help with the accuracy of the data derived from the test (Dumas, 1993, p. 123). Deciding how many people should be in the testing group can be different depending on the site. Jakob Nielsen in a usability test found that not quite half of all major usability problems were detected with three participants. He also found that 80% of the usability problems in a website were detected with between 4 and 5 participants. Any additional participants are less likely to reveal further significant information but are used just in case (Dumas, 1993, p. 127). An absolute minimum of people for any test of a site should be about three people.

Test Labs

Usability tests are sometimes conducted in special usability laboratories and some large companies even have dedicated rooms for this purpose (Nielson, 2000, p. 339). The Maya corporation based out of Pittsburgh, PA, use two testers and six potential users of a client's site. They conduct their entire test in a lab and claim that a formal test will take about a week to complete. Included in their test is a complete report of the test, a video and analysis of the users and even a detailed guide of suggestions for your site. All of this depending on the detail of the test costs between $15-35k (Maya, 2003). There are other settings testers use to conduct usability tests. International usability testing lately has been the key to building great sites. Very large corporations have usability laboratories in multiple countries. Some testers even carry portable test labs to allow them to travel to many different places easily.

International Testing

International usability testing has two great characteristics that stand out over local testing. First, the user may be speaking a foreign language and will not use the site if they cannot understand the information on the site (Nielsen, 2000 p. 339). Secondly, performing tests in countries with cultural differences can be useful information when finding out how useful a website is going to be. Users from around the world will use a website differently. Those who design their sites solely for a single country will be turning away more than half their customers. Tapping into the international market will be an advantage over other websites.

Characteristics: Good Sites/Bad Sites

Good websites, as mentioned above, allow the user to “surf” and perform functions easily in order for users to complete tasks. There are many parts to a website that can give it a high level of usability. Headings in a good website use many, carefully selected headings, with names that conceptually relate to the information or functions they describe. Headings provide strong cues that orient viewers and inform them about a page's organization and structure. Headings also help classify information on a page. Well-designed headings are an important tool for helping users scan text. Write headings and page titles that clearly explain what the page is about and that will make sense when read out-of-context. Headings are often removed from the context of the full page and used in tables of content and search engine results. This means that the headings should clearly tell users what is at the other end of the link (usuable, 2003). For home and navigational pages in a website, use short pages. Pages need to be easily navigated and it saves time and effort on maintenance. When deciding on the layout of the site makes sure there is a high to low flow of information on your site. More important information should appear first so user can find it quickly. Being consistent from page to page with logos and graphics. This will greatly help users to stay focused on completing their tasks especially users over 65. Reducing unused space reduces the need to scroll and search unnecessarily. This will overall help users accuracy and speed when surfing. As with putting the most important information at the top of the sight, do the same with links. Make them clear and easy to distinguish. In addition, make the links descriptive enough to prevent the user from going where they didn't want to go. This will cut down the users time and make them less impatient. Avoid using unnecessary graphics (Spool, 1999 p. 86). Use graphics that lead to a better understanding of the information being presented. Making navigational aids easy to find and consistent will greatly increase sites usability. It's fun to use colors but in most cases excessive use of colors will inhibit the user ability to surf effectively. Color coordination is the way to go in terms of navigation and different topics to be presented. Bad websites will lack in any of the areas mentioned above. Bad websites may have mostly great traits but lack in an area that will cause a user to quickly leave and render your site useless. Navigational obstacles are one of the biggest turnoffs. Even before that when user log onto a site, the size may be the culprit. If the size of the homepage is too large for a quick loading, users will simply not wait around for it to load. Things like graphics and tickers are big offenders.

Revealing Clues

There are a few clues to look for when finding out how usable a site is. If there is lots of traffic and few purchases obviously there is a problem. If users are only hitting the top-level of the site and not reaching the lower levels there are severe navigational problems. One of the goals to when designing a site is to design a website that creates a desire for a surfer to come back. An obvious clue to a bad website is the lack of repeat visitors (Usableweb, 2000). In general, websites should be simply allowing users to accomplish their goals in a timely effective way. Users will enjoy the site more while building a rapport with the site. As a result, this will increase the sites reputation and traffic.

Conslusion

With the amount of websites on the World Wide Web virtually multiplying every year, making a website as usable as possible can greatly increase the chance that a site will stand out over the others. Usability testing can reveal incredible things or allow the designer to discover terrible things about the site that he or she overlooked. Usability testing is something that is constantly evolving. It follows right along with evolutionary changes that are expected in humans and technology. There are only a handful of websites that have a high level of usability. My research has found that most websites are designed with great appearances and impressive features but their designers tend to forget about the most important factor: the users.

References

Dumas, Joseph (1993). A Practical Guide to Usability Testing. Norwood, NJ: Ablex Publishing, 4-123.

Nielsen, Jakob (2000). Designing Web Usability. Indianapolis, IN: New Riders Publishing, 10-341.

Krause, K. (1996). Usability and Website Testing. [Online] Available: http://www.cre8pc.com/ui.html , (N.D.).

Many Athors. (1996). Methods. [Online] Available: http://usableweb.org/topics/000862-0-0.html (March 5, 2001,February 26, 2001, February 23, 2001).

Spool, Jared (1999). Web Site Usability. San Francisco, CA: Morgan Kaufmann Publishers, 10-111.

Website (2003). Usability and Website Testing. [Online] Available: http://www.maya.com/web/who/who.mtml , (N.D.).