Creating Research-Based Web Sites with Netscape Composer

Testing Your Site

It is very important to take the time to test your site.  For the assignments in this course, it is expected that every link on every page be working.  It is also expected that the information on the pages be clear, readable and accurate.  To guarantee this level of functionality, you should test your site in several browsers and on PC and Macintosh platforms.  Also, change the size of the browser window on the screen to see what the site looks like when viewed in small or large windows.

Once you have tested the mechanics of the site, you may want to test the “Usability” of the site – the ability of your users to access information that they are looking for.  Web Site Usability Testing can be conducted in a very simple fashion – here is an outline:

1. Identify a test subject – someone who is included in your defined “audience” is a preferred tester.  If you design a site for middle school students, you should have middle school students test the site.
   
   
2. Create a short list of items that you want the tester to find on your site.  For example:  “Where would you click to locate information on the history of my invention?”  Be sure your list of questions is inclusive of the different areas or types of information you want your users to access.
   
   
3. Seat the tester at a computer where the home page for your site is opened, and ask them the questions you have compiled, one at a time.
   
   
4. Make the tester feel at ease and let them know it is OK if they are unable to successfully find the information you want them to look for – let them know that the reason for the testing is to find the problem spots in the web site.
   
   
5. Observe what the user clicks on in response to your questions.  Make notes on items were users don’t seem to know what to do or where to go.  Also note the times where they go to the wrong location to get information.  Give users plenty of time to try to answer your question. It is important that you only observe and not give hints or tips.  You will not be seated next to every person as they access your web site, so do not help your testers either – this will give you the best information on what needs to be fixed.
   
   
6. Repeat this process with several other users if possible.  More users give you more reliable data on what works and what doesn't.
   
   
7. After testing, compile the results from your testing into a ranked list of recommended revisions for the site.  Items where all of your testers had problems should be ranked higher than areas where only one or two users had difficulties.
   
   
8. Revise your site based on your testing.
   
   
9. After your revisions, test again!
   
   

Of course, you may not have time for testing over and over again, but this is the process that professional web designers use to evaluate the success of the presentation of information in their sites.  Even if you just test one time with one person, you will gain valuable information that lets you know if you have been successful in the information architecture of your web site.  Remember – a easy to use site doesn't happen by accident.