Usability testing: Articulate features

After the product development, usability testing is a necessary link, and before testing, we will generally go over the software to familiarize ourselves with the product, and the problems may be encountered in the heart there is a general judgment.

The author of this article @uprit recently made a usability test for navigation products, the main process is to allow some users to be accompanied by the host to complete a series of typical tasks have been designed to discover some of the software usability problems.

The whole test came down and found a more interesting thing: there was a task that looked more complex (group navigation) but had a higher rate of completion, while one task seemed simple (search around) but the rate of completion was low. This is a bit different from what was expected before the test.

Let's take a look at the process of these two tasks and the interface involved. Before that, let's introduce the two features.

The first is "periphery". The whole logic of "periphery" is relatively simple: first identify the center point and then search the various locations near the center point. In short, it is "fixed-point –> search", as shown in figure:

This, because the mobile phone has positioning function, "fixed-point" and more is our current position, so, to the user, the fixed-point is generally done automatically by the machine, the user needs to do is left only "search". The process is really simple.

Then there is "group". The so-called "group", mainly in many people driving out, you can through the "group" to share the route, real-time navigation, and see each other, in addition to a number of additional functions, such as chatting. The whole logic of "group" is a little complicated: first, you need a person (group master) to create a group, after the creation, need him to tell the other group name, password, this process is generally through the SMS invitation to achieve, received the message (members) after getting the group name password, enter the navigation, search the group name, enter the password , join; then the group master planning route, after planning, members can see the shared route on the map and can see each other's position. This is illustrated below:

This process involves a number of pages, at the same time requires more than one phone to interact with each other, the entire flow slightly more complex.

However, the completion of the "group" rate is better than "peripheral", which was not expected before the test, the completion of the two tasks are as follows:

(green: one complete; yellow: many attempts; red: Failure)

Why does this result? I try to find out why. Let's take a look at the interface design from the point of view.

First, we look at the "perimeter" of the interface, "peripheral" page, the top is the center point, the following is the search box, then the category, and finally the virtual keyboard. Enter the page by default the search box is activated.

What's wrong with this page? I think the main problem is that the relationship between the center point and the search box is ambiguous. For example, search the restaurants around the "Tianzifang Square", the task failed or many users tried, they will now search the box search "Tianzifang Square", and then back to the "surrounding", click "Restaurant." The result here is not the restaurant around "Tianzifang Square", because the user has misunderstood the setting of the center point.

In addition, the location center point is also a more redundant function, mentioned earlier, most of this positioning process can be automatically completed by the phone, then no need to set the user. In the actual test, the user basically did not notice that the position was clickable and set (of course, they completed the task through other paths).

Secondly, the search here, is not clear meaning, do not know whether to search the site or the type of search. In terms of hierarchy, the search should be a hierarchy of different types, but the interface places the search and location center in one piece, which is also a cause for confusion.

Finally, after entering the page, the default search activates the input box, as if to prompt the user to enter text, in general, the user needs to find a thing, if he has a clear goal, he will tend to use search to find, if the goal is not clear or clear target and clear categories, he will use the classification. For the "search restaurant around Tianzifang" This message, it is clear that "Tianzifang Square" is a clear goal, "restaurant" is a vague goal, so users may directly use the search to locate the center point, which will produce the wrong results.

Let's take a look at the entire "perimeter" process and the pages involved:

This process does not look very complicated, and if you are familiar with the operation, it may be quicker to operate, but, as such, it gives the user a great degree of freedom, so that there are multiple operations on one interface that may lead users to stray from the main task and ultimately confuse the user. , the confusion caused by the misunderstanding of the interface also leads to the decrease of the task completion degree.

Next, let's look at another task, "group" navigation, and the following figure is the process of "group" navigation and the pages involved:

"Group" Navigation is a complex task, the main face of two types of people, the group and members, basically one step interface, although cumbersome, but each interface to convey the task information is clear and clear, to the user's freedom is very small, users in the complete process before the completion of the current task is almost difficult to jump out, so, The whole process is the same as installing the software, always "next" until the final "finish".

Through the process comparison of these two tasks, we find that if you can't make a task process simple, do it as much as possible to make it clearer, less on the interface to give users more freedom; If a task is not complex, as little as possible in the interface to the user confusion, while, some functions (such as set the center point), If not necessary, cut off and try to make the operation more concise.

Above, is from the process aspect to analyze. When talking about the problem with the circle, she gives another angle that is more interesting: the user's point of view.

In contrast, "perimeter" is a more commonly used function than a process, even some users may not touch the "group" for a lifetime, the "group" for users is a strange thing, then in the usability test, the user to complete a strange task, especially when the task is slightly complicated, The user assigns more attention, has already had a heart anticipation before doing so, also has the patience to complete this task.

So, for the "perimeter" of this common basic function, users in the implementation of the task, may be more according to their existing habits or intuition to operate, will not allocate too much attention to the software tolerance is not so big, once a little confused, waiting, or wrong, users will complain higher.

In the Kano model, if the basic needs are not met or poor performance, the user's dissatisfaction will increase sharply, and such needs are met, can eliminate customer dissatisfaction, but not to bring the increase in customer satisfaction, in contrast, if the demand for charm once satisfied, even if the performance is not perfect, Can also come to a sharp increase in customer satisfaction, and such demand, if not met, often will not bring customer dissatisfaction. In this navigation product, "periphery" should belong to the basic demand, "group" belongs to the charm demand, the actual test result also is the user to "group" overall satisfaction to be slightly higher than "periphery".

It is thought that, for the common functions, must do a good job, not commonly used functions, must be clear. Whether the task is simple or complex, the primary satisfaction is clarity. If you can think of a cooler way, that's fine, but try to make sure that's the premise.

Author blog: Winter Source Zhi

(Lei Feng Net warlial an article, reprint please indicate from Lei Feng Net and author)