How to Conduct a Heuristic Evaluation

Jack Nielsen, usability guru, came up with the method known as heuristic evaluation. The goal is to find out how Internet users interact with a website. Steve Krug, author of "Don't Make Me Think," emphasizes that testing usability doesn't require a random sample of hundreds of users. Even a few users is better than none.

Instructions

Difficulty: Moderate

Things You’ll Need:

• Private space
• Desktop computer
• Camcorder or video recorder

1. Decide who your target audience is. Place a notice on a college bulletin board if you're targeting 17- to 24-year-olds. Consider placing a flier at the library, local cafes and markets for an older audience. Offer a monetary incentive. Find out what consultants just starting out are charging per hour in your area. Pay that amount for older users, half that for students.
2. Set up a private space to conduct your usability testing where you won't be disturbed. Fast web access through DSL or cable is preferable. It's best to use a desktop computer because that's what's familiar to most users.
3. Set up a camcorder or video recorder in an unobtrusive place where it can capture the screen. This is important because you don't want to rely on just your tester's observations. Get another colleague to sit in on the sessions if you can't capture them on video.
4. Ask the subject some background questions before you start the heuristic evaluation. Find out what their occupation is, how many hours a week they spend on the Internet and what sort of information they typically look for online. This will help loosen them up.
5. Test your homepage first because it's your most important page and it gets the most hits. Let users figure out for themselves what the company is about. Encourage them to think out loud and be honest with their criticism.
6. Give your users a specific task. Identify what you want people to do on your site once they get there and ask subjects how they would go about doing it. You want to see how hard it is for them to accomplish this.

Tips & Warnings

• Krug says repeat testing is more important than the number of users. Conducting three heuristic evaluations on three users is better than conducting one on nine users.
• Do what you have to do to make the setting as comfortable as possible for your subjects. You don't want little inconveniences getting in the way of their activities.