Simple and quick usability testing

Usability testing is one of the best ways to improve your product, which is an indisputable consensus within. Just because the research staff is always insufficient, so in order to enable colleagues in various departments to carry out a number of studies and testing work more quickly, we have collated a number of simple documents and tutorials, and plans to focus on training to popularize some of the user experience methods. Therefore, it is particularly emphasized that the test method presented in this article is simple, informal, small sample, to identify serious problems for the purpose. If you want to learn more about how the test works, see the following books:
Handbook of Usability Testing

Usability testing and the

Measuring the user Experience

Before I begin, I would like to stress that testing starts as early as possible and can be done with simple equipment.

What is http://www.aliyun.com/zixun/aggregation/8135.html > Usability test?

Usability testing is not complicated, but a simple generalization is to observe the user's use of the product. If we expand it a bit, we can interpret it as: by looking at a representative user and accomplishing a typical task for a product, we define usability issues and solve them. The goal, then, is to make the product easier to use.



How to design and conduct usability testing

Based on the above definition, we will mainly answer the following questions in the section on how to design and conduct usability testing:

• How to design tasks

L How to find Users

How to test

How to analyze to find usability problems

How to design test tasks;

Design test tasks are at the heart of the Usability testing initiative, and we recommend that you determine the criteria for user recruitment after the task is designed, as the tasks involved in the test are directly related to the user experience you are describing. The former is confirmed, and the latter is clearer.

Before you design a task, you need to ask yourself repeatedly:

"Did the test task I designed really reflect the user's actual goal (not what I think the user wants to do)?" ”

After asking more than three questions above and confirming that you can answer for sure, you can start designing tasks.

The usual approach is:

A lists a list of tasks, using simple phrases to describe the tasks involved in the test-primarily for internal personnel. Due to the quick test, therefore, the task should not be too much, must be important, the core, you feel may have problems with the task;

b After sifting through the task list, turn the task into a scene--the scene is what you want to read to the user or to the user. So you have to include the user's goals and motivations--because for the user, your function is not important, it is their purpose and the process of accomplishing their goals. At this point, you can ask your questions again.

C Determine the requirements for operational tasks: for example, whether to need a new account, is not required to prepare the necessary documents and so on;

D Prediction Test: Predictive test is mainly to find out the problem of task design, you can find colleagues in the company, use lunch time to quickly complete

How to find Users;

Before we use all kinds of resources to find users, we need to be clear about what kind of user we are looking for.

What kind of people are you looking for?

Previously, in the test planning phase, we communicated with the product and design colleagues, their description of the user, often: 25~30岁, office workers and so on. But in fact, we are most concerned about the user's behavior in the test, therefore, in determining the user screening criteria, the most should focus on product experience and use behavior, rather than demographic characteristics. For a more detailed explanation of this, refer to this article. http://www.uie.com/events/uiconf/2008/articles/recruiting_participants/

How many people are you looking for?

When it comes to finding a few people to test, you have to ask Nielsen's classic chart. Although academia has a lot of debate about whether 5 users are enough. But from our point of view of practice, as long as the first phase is well-defined and the right person is found, 5 of users can really find the obvious usability problem, and of course, again, the goal of the quick test is to identify serious problems rather than all of them.

In addition, we found that in normal tests, in the observation of the first three users of the test, the product and design after a single mirror is often focused on, writing fly, but to 5 after the new information less and more, we are not in a daze, is simply sleep (like the rabbit in the picture). So, from the point of view of working state, 5 test users can ensure that they are focused and willing to observe and listen.

How to find a user?

After determining the criteria for screening and the number of users, finding users is a headache. Here, my advice is, because it is a quick test, try to use all the means, whether it is to find colleagues (colleagues in the same department do not find), friends, friends, web forum ads, etc., as soon as possible, and in the recruitment, insist that we can "unscrupulous". Of course, it would be nice to have a list of users that you normally maintain.

How to test

Because it is a fast test, the moderator from the use of research staff into the product and design staff themselves. So the point here is to not try to teach a user how to use a product in a test, or try to sell your product to a user.

What does the host do?

What does a recording person do?

Whether it is conditional real-time observation (with a special experience room or tool), or need to look back at the video to observe, note that the focus of the record is not what the user said, but how users use. Remember, in testing, what you do is more important than what you say. Of course, because it's a test for serious problems, and serious problems are always obvious, you can record the problem at the same time, but don't rush to discuss the solution to the problem. Because immediately think of the solution or the user's proposal is not necessarily the best, this work can be left to be quiet thinking or when we discuss the

How to analyze to find availability problems

After the test is completed, the moderator and the observer need to quickly organize the useful information while they are still fresh, can use the convenience paste, can also specially empty a whiteboard or create a document. In short, the user-related operations, questions raised, and our own found problems quickly write out. But don't jump to conclusions.

After all the tests have been completed, organize the existing post-it, list, and so on, and then identify the most serious problems and fix them quickly. At this point, the focus is on being clear again and again that those problems are the most important and can be repaired immediately. That way, the test results are executable, not just a list of issues that are archived.

Source: http://uedc.163.com/4151.html