Home > Developing > User Testing

Developing: Testing the Web site

Testing your Web site brings to surface any problems, confusion or annoyances users have with a specific task. Look for the underlying patterns in what people say and how they react. Don't try to follow their desires verbatim.

• What to test
• Who to test
• How to test

What to test

There are several components of a Web site that can and should be tested. They include, but are not limited to:

• Clarity of content: Is the text easy to read and understand? Is important information clearly displayed and identified?
• Consistency: Are similar elements throughout the site presented the in same way and in the same location?
• Labeling: Are links and buttons descriptive enough? Is it obvious to the target audience what information the links will provide and where they will take them?
• Visual Presentation (look and feel): Are the colors and images appropriate for the subject? Are pages easily recognizable as being part of a specific section? Is there a clear distinction of global navigation to local navigation?
• Overall Use: Was it easy to get around in the site? Are there navigational methods to help the user know where they are, where they came from and where they are going? Was the task completed with little to no problem? Are there other elements that could have provided better, quicker or more satisfying results? Would the user visit this site again?

< top

Who to test

Testing with actual targeted users of a site provides three benefits:

1. You find out what your users want before you complete your site.
2. It puts the user in control of how a task is best organized and written rather than a biased non-user's.
3. It provides valuable data for baseline measurements of success and failures for return on investment and future enhancement priorities.

< top

How to test

Knowing how to test is the single most important aspect of user testing.

Know this first

• Know what you want to test. (see what to test)
• Know what you will ask.
• Recruit participants.
• Know how long you have to test and compile the results to implement changes.
• Decide the type of testing you want to perform

Now do this:

• Set up an area to conduct your testing.
  • Provide a computer (or laptop) with Internet connection.
  • Position a mirror behind the computer to get a full view of the participant's face.
  • Place a video camera in the room to record the test.
  • Get a stopwatch to keep track of time during tasks.
• Establish facilitators
  • Primary person to conduct the test
  • Secondary person to watch, listen and take notes on what participants say, where they look, what they click, their body movements and expressions, points of frustration and confusion, etc.
• Conduct the test
  • Provide a detailed instruction guide to the site you want to test.
  • Make sure participants know they are helping by trying out the Web site. The site is being tested, not them.
  • Get participants to think aloud as they work.
  • Let participants express their reactions.
  • Listen! Do not lead. Be sure to stay neutral in your words and body language. Be careful not to ask leading questions that may skew the participants' responses.
  • Take detailed, useful notes concentrating on observations of behavior rather than inferences.
• Compile the user data
  • Find the patterns with all test participants.
  • Consider influences that might skew a user's session (foreign room or computer, being observed, desire to not mess up, etc.).
  • Organize findings into three levels: "CRUCIAL," "NEEDED" and "DESIRED" with impacts and recommendations for all of them.

< top