Usability testing doesn’t have to be expensive. It doesn’t have to be time consuming. The most important thing is to actually do usability testing, and of course to respond appropriately to the feedback from the test.

Studies show that you really don’t need more than five people in a study. If all you can get is ONE person, that’s better still than not having any. However, having 3 to 5 people who have not been involved with your project, test your website’s usability, will yield really solid results.

If you’re like me, you’ve probably got at least five people that are also web developers or in similar fields that you can rely on for feedback. Use them as a sounding board for your first review of a project. Make it clear that you want this to be a trade out; you’d like their honest feedback in exchange for your doing the same for them in the future. Ask them to tell you what they think the primary and secondary purposes are suppose to be for the project — don’t get them the summary of the project ahead of time. You’ll get better feedback this way.

This group is very likely to give you good feedback, but they will likely interact with the site in expected ways and not as the “typical” user would. I’ve also found using other web developers for a round of simple usability feedback to have a high success rate for completion. Web developers get the importance of this.

If at all possible though, you need a sampling of people that are not connected to you or your project. Often you can get this sample of users from your client - set this up in advance. Ask the client to provide you with email addresses of five people that you can share the project with and what you plan to do. If you’re clients are like mine, they have likely disagreed with some of your recommendations about information architecture. I have found that a review like this really does a great job of validating your previous (often ignored) recommendations.

Ask these test users to tell you what is the first thing they notice on the site. Keep that question vague. Let them tell you what they see. If you can, watch them in person as they interact with the site. Have them tell you what they did first, and why. What problems did they encounter? Ask them to find something specific on the site. How did they get to it, if they indeed could find it? The answers will reveal a great deal of solid data for you.

The cost of doing a study like this is really next to nil. Share the feedback from the users with your customer, but try to avoid doing a formal write up. Why? The more time you spend on the testing, the less likely the client will be willing to engage in more usability studies down the road. Keep the costs down so that there isn’t a barrier to doing them. Usability studies are far and away the most important thing you can do to improve a site’s design and function.