How do you find out about what the user wants? Or if your application is effective?

You should conduct research, but it pays to take a critical approach and find the best method.

"It's really hard to design products by focus groups. A lot of times, people don't know what they want until you show it to them." - Steve Jobs, BusinessWeek, May 25, 1998

There are lot of techniques out there for research, and just like almost everything, there’s no "one size fits all" approach when it comes to determining what users really want.

What are these techniques? And when do you use them? [Disclaimer: These examples are just a starting point].

User Interviews:

What: One-on one-interview with selected users

When: A project is on a budget, with no logistics to set up more elaborate research. Done at the very beginning of a project.

How: You need to create a list of questions that will pinpoint the following:

• The user's experience with the brand and/or competitor
• Goals/Needs/Wants that the user seeks in your site
• What does the user do after visiting/using a site?
• Does the user ask for help from colleagues/friends while doing a certain task?

Tips:

1- Avoid biased questions at ALL COSTS:

Bad: How did you like the search engine?
Good: What do you think of the search engine?

Bad: Do you like this feature?
Good: How would you use this feature?

2- Never give out answers such as: Excellent/Great/Good. Always remain neutral. "Ok" is a fine answer if you don't want the user to elaborate more.

3- Always look out for body language and non verbal communication: You will encounter users whose body will give a different answer than their speech.

4- Always get the users to sign a release form, granting you the rights to use their answers.

Contextual Inquiry:

What: Observe the user doing certain task in his own environment. This is very informative.

When: Information about users is scarce, or tasks are fairly complex, and you have a targeted environment (banks, universities, schools).

How: Two types of observation: active and passive.

In active observation, the user actually plays the role of the teacher, showing you how to do certain tasks by explaining the process and goal for each one. This gives you insights on why the user is doing tasks in certain ways.

In passive observation, you ask the user to act as if you are not there. You want to observe the user's behavior in as natural a setting as possible. Essentially, you pretend you don’t exist. Chances are if the user needs to ask for the help of a friend or fellow site user, you will see him doing that naturally.

Surveys:

What: Close ended answers, answered online or offline, helps to discern certain patterns.

When: Whenever you need quantitative results (e.g., 74% of college students use twitter).

How: You will have to set a series of questions that require no speculation. The easiest forms for analyzing results are Yes/No and multiple-choice questions. If you are planning to do some usability tests, surveys can also help you form an idea and determine how to proceed. To assess usability, you will need to ask attitudinal questions (Strongly Agree/Disagree).

As you have noticed, surveys can be used as a pre-step also for other research methods, such as helping to define user groups, further questions, tasks, etc.

In the next post, I will cover the three remaining techniques: card sorting, focus groups and usability testing. Always bear in mind, I am merely scratching the surface, and never forget; search engines and books are your best friends.