UX Evaluation methods

There are different ways to evaluate the success of your website.

Normally in this class the evaluation has turned into creating a survey and sending it out and collecting the responses, but before we do that we should think: is it the best way to evaluate my website? can I really learn anything by sending out a survey? how much time are my classmates going to spend responding my survey?

There are other evaluation techniques you could use. Here are a few. You will need to decide what method you want to use and justify your choice.

1. Focus groups

A focus group is a moderated discussion that typically involves 5 to 10 participants.  Through a focus group, you can learn about users’ attitudes, beliefs, desires, and reactions to concepts.  Focus groups are group discussions. You recruit several targeted users, get them in a room together with a moderator, and ask them questions.

According to Cardinalpath, focus groups are wonderful for:

  • Learning how customers feel about your products, including their past experiences with them.
  • Generating ideas for new products or features (or getting reactions to your ideas).

For example, let’s say Facebook wanted to identify new features their users might like. Holding a series of focus groups would be a great way not only to get feedback on ideas you’re thinking of implementing, but in generating ideas you’d never even thought of.

Focus Groups are best used early in the process, i.e. before development (or re-development).

Pitfalls of focus groups include:

  • You only get subjects’ stated opinions. Sometimes, what people say is very different from what they’ll do. (Famously, the Edsel tested very well in focus groups.)
  • Participants may influence each other. Too often, one or two opinionated individuals will dominate the group. Other participants – who may have better insights – fear to differ.

You can see what a good and a bad focus group is here.

2. Usability Tests

Usability tests aren’t groups ; they’re conducted one-on-one. And you don’t simply ask user’s opinions on your site. Rather, you to observe how people actually use and react to your site.

In a typical test session, a moderator sits beside the user and assigns a series of realistic tasks. The subject is asked to “think out loud” as he completes the tasks. The moderator carefully observes what the user is doing, and can ask follow-up questions, etc.

Advantages of Usability Testing Include:

  • It can (and should) be done at any stage of development: from wireframes through to live sites.
  • It is the perfect complement to analytics: Analytics tells you what is happening on a website; user testing tells you why.
  • Recorded sessions are great for convincing doubtful or reluctant members of the development team that the site has problems that need to be fixed.

The difference between focus groups and usability tests is that to prepare for a focus group you need to prepare a set of questions, however for a usability test you need to prepare a set of tasks you want them to do.

3. Online surveys

Online surveys are a quick and incredibly useful tool for gathering all sorts of user feedback. The important thing when planning your survey is what kind of questions do you want to ask. Open questions will give you more information than yes/no answers.

Remember than when planning the survey, the questions need to come out directly from the specifications/goals you defined at the beginning of criteria B.

Uxforthemasses  gives you some recommendations on the kind of questions you should include in your survey.

4. Other methods

Sometimes other methods are more appropriate: Surveys, interviews, card sorts, eye tracking studies, remote testing, automated online tools… But one-on-one usability testing will always be the usability practitioner’s most powerful tool. There’s simply nothing better than actually watching someone use your website, and being able to ask live follow-up questions.

Card sorting. Card sorting is an interesting method to test your web tree.

First click testing. Where does the eyes of the user go first?