Friday, November 11, 2011

Active observation: lean forward, write lots

User research is a lean-forward activity: you have to remain actively engaged. It's very different from watching TV: a lean-back, passive activity. 
Good user observation techniques require you to constantly note down what you see and hear, and formulate questions that you'd like to ask. If you sit back for even a minute you lose the flow of the user's task and your mind can start wandering to solutions.

Set clear goals

The key to active observation is to go in to the session with a clearly stated goal or set of goals.
  • What tasks are you there to observe?
  • What interactions are of interest to you?
  • What actions and behaviors are out of scope?
Usability studies have pretty well defined goals - you have a list of tasks and you are interested in how well the interface you're testing supports the user in completing those tasks.

Site visits and field studies normally have broader goals - finding out how work "happens." In these situations you have to clearly identify what parts of the participant's life you are interested in following.

Capture key data without judgement
The most valuable data you can record is a running description of the user's quotes, and observations of users' behavior and actions. The output should read almost like a movie script.

Try to refrain from making value judgements about the things you see users doing. Instead, just be sure that you have reached a proper understanding of why the user thinks they are necessary.

It helps to think of yourself as an understudy to the person you are observing. If you had to act their role, what would you need to know?   

Leave solution-type thoughts until you're back in the office. Time with users is valuable, and if your mind wanders to solutions you'll curse your inattentiveness later on when you can't answer questions about the task you observed.

Take hand-written notes. Even if you're a fast typist, it's hard to be non-linear on a laptop and the keyboard noise is intrusive.

Let the user do the talking
In usability sessions it's best to use a think-aloud protocol, where the user gives a running commentary of what they're thinking as they work through the tasks you've set. If they fall silent and you know they're still working through an issue, you can prompt them to remember to think out loud for you.

In a site visit, you have less insight into what's going on inside the user's head. It may be necessary to ask questions to fill in some gaps, but it's usually best to leave the questions until a natural break in the work.

Leave the video camera at home
A video camera might initially seem like a good thing to take on a site visit, but it can be more trouble than it's worth. The problem is, you'll be less likely to take good notes because you start to rely on the video camera capturing everything. Of course, you will never find the time to watch all those hours of recordings so it's better to capture the information in real time as you are sitting next to the participant. Video cameras can also freak participants out and make them behave differently.

The one time that video can be useful is during usability studies when something goes wrong and you're not sure what it was that a user did with the software. Normally a screen recording program is sufficient to grab the information you need. Don't plan on re-watching the whole interaction at any point, you probably won't have time. 

Do bring a still camera
On site visits, take pictures of artifacts. An artifact is anything that your participant uses to help them do their job. It could be the typed up list of phone numbers they keep in their wallet (even thought their phone has an address book). It could be the post-it note on the side of their screen with the steps they need to take for a difficult task. It could be the crazy form they have to fill in every time they submit a work order.

Always ask permission before taking pictures, and wait for a break in the flow wherever possible. Sometimes you might need to blank out confidential information before you take the picture. 

Later in the design process you'll use those pictures of artifacts to remind you what data is important, what things people have difficulty remembering, and what physical items need to be included or replaced in your new design.

Observe in pairs to catch all the points
You will miss things. Either because you're busy writing something down or because you just aren't attuned to certain behaviors.

Having two or more observers gives you backup for the important points and has the added benefit that different people on the team will have awareness of different aspects of the interaction. One person might innately grasp the technical issues, whereas another might be good at reading body language and implicit emotions.

Limit your time doing active observation
Active observation is tiring. Your writing hand will cramp up. Your brain will hurt. To start with you're unlikely to be able to focus for more than about one to two hours. You won't want to schedule more than two observation sessions in a day.

As you get more practiced at listening and taking notes, your stamina will increase. Soon you'll be able to run four to five one-hour usability sessions or user observations in a day. You'll still go home drained at the end of it though.


Creative Commons License 
RSS  e-mail


No comments:

Post a Comment

Please keep your comments respectful, coherent, on-topic and non-commercial.