What does a modern usability laboratory look like?

Many people imagine a usability lab as a space where product prototypes are shown to people, everything is recorded and changes are made to the interface based on the data obtained. Yes, it is in the laboratory, but at the same time it is only the tip of the iceberg. Directly testing is already a spectacular finale. Before the development of the first prototype of a product starts, you need to do a lot of different things.

A usability laboratory is a few rooms in which different contexts of using Beeline services are modeled. We have three zones: office, home and cafe. All three zones are equipped with recording equipment, which allows specialists to accurately see and record what the user is doing. In all three zones, the “windows” come out of the hall where the observers sit - for the study participants they look like mirrors, for the members of the working group who came to observe the study, they are transparent.
')

Laboratory premises

Another laboratory room is a large screen conference room where presentations are made based on research results. In it, there are design teams for joint development of design solutions. If necessary, the “Cinema Hall” test zone is created in the same conference room.

Zone "Cinema"

Researchers and observers see the screens of users, the image of the face and voice of the respondent is also captured and transmitted. In parallel, with the help of a special infrared camera, the movement of the eyes of the research participant is recorded. During subsequent processing, you can view the "heat map", on which visible and "blind" interface elements are clearly visible.

Aytreker

He is on the side

The entrance to the laboratory is located up to the turnstiles of the office security.

How is working with the new product

We connect to the process of creating products at the earliest stages. At the very beginning, we get an idea formulated literally in the form of two sentences.

Work begins with the analysis of user tasks. First of all, you need to determine who will be interested in the product (for example, if we are talking about a loyalty program, then these are people with discount cards, if we are talking about parental control, these are parents whose children have the ability to independently access the Internet, and so on). Several representatives of this audience are invited to the laboratory. A kind of brainstorming begins, the main purpose of which is to understand what people want from a ready-made solution. There are special methods that allow you to get enough accurate wishes from the audience. At the end of the discussion, the respondents select the most important tasks for them.

Usually in one study take part from 5 to 9 groups. Knowing the frequency of mentioning tasks in different groups and assessing their importance, we can calculate the priority of each task. After that, the project team can determine what will necessarily be included in the first release of the product, and what will be added in the next stages of development. ”

Very much depends on the skills and personality of the moderator. The moderator must be able to behave successfully and switch between several professional roles in time. First, he is a hospitable host who can establish emotional contact with a respondent or a group, build trust, relieve stress, and do everything possible so that participants say “with their voice and what is important for them”. Secondly, the moderator is a strong leader who is responsible for everything that happens during the study. He must quickly and in an organized way respond to all problems that arise, control the pace and quality of work of the respondents. Thirdly, the moderator is an impartial observer who tries to contribute to the results of the work of the participants as little as possible of his own experience, opinions and assessments. And finally, the moderator is fully responsible for fixing the research data for which, in fact, the laboratory is organized. As a rule, researchers come to us from professions related to social studies - psychology, sociology, marketing research. There were experiments with moderators from the technical divisions, but they often do not understand how you can not understand the work of the product. Where the psychologist sees the problem in the interface, say, software, the technical specialist sees the problem in people who cannot understand something extremely obvious to him. That is why the best researchers are people with liberal arts education or former interface designers.

Data from users is not the only source of information about the functionality of the future product. In addition, we conduct a study of global and industry practice, most often there is the opportunity to look at an analogue in the Western market and take into account already known problems.

In parallel, data is requested from experts in the field within the company or from suppliers of technical solutions. Thus, information from customers is verified and updated.

How client requirements are clarified

The list of tasks to create a product interface is often not enough.

If a product turns out to be functionally or content-loaded, as, for example, it happens with corporate portals, you need to understand into which sections the content should be broken up and how to name these sections so that customers will not have problems finding information in the future. In the creation of information architecture helps card sorting method.

During the group discussion, several people must agree among themselves and explain to the moderator which groups they want to break up the content offered to them, by what criteria the selection of groups is carried out, as the resulting groups would like to name. There are special analytical tools for processing this kind of data, their principle of operation is to take into account the “distance” between individual content units, it is calculated which of them often end up in the same group and which always diverge into different categories.

From the names of groups proposed by customers, you can choose the most successful. Subsequently, this will not only facilitate the search for information, but also, for example, will help SEO, because the section names are formulated in the language of customers.

There is a card sorting

In some cases, other studies may be required. For example, we find out what steps and in what sequence customers take in their normal life if they face a specific task. You can get this information in different ways, including by listening to several dozen calls to the call center.

The research results are taken into account when creating prototypes. For prototyping web interfaces and software, standard tools like Axure Pro are used, prototypes for mobile devices are made using software specially developed for this purpose, and paper prototyping is also used in some cases.

All these methods pursue one simple goal - to have detailed, user-friendly requirements tested on real clients prior to the start of development. It is known that the later the problem is discovered, the more expensive it is to fix it. It may feel that research is delaying the project, but in fact there is no better way to avoid the numerous rework and "war of opinions" that accompany the creation of products in any large company.

The later corrections, the worse.

Design

Usually, up to three design iterations are performed, each of which is verified on users by means of usability tests. This is the most famous and familiar stage. Here it is important to check not the quality of individual buttons, filters and pages, but the success of user tasks. Tasks are set for the respondent in the most neutral, natural, simple manner. For example, a task can be formulated as follows: “Find the tune of Peter Nalich“ Kam that May boudoir ”, order it and set yourself instead of waiting beeps”. The option is not very successful, since we suggest the respondent a sequence of steps. Much better would be such a wording: “You called a friend and heard this melody (we give the number to which the dial tone is set to the song Nalich, we don’t call the performer). Make sure that you play the same melody. ”

During the tests is the study protocol. All the words and actions of the respondents are recorded and processed. Data is collected on the success of the work, time spent, the emotional assessment of what is happening, as well as problems that arise.
For problems, ratings are calculated depending on their frequency and impact on the result. The results of the work are graphs of efficiency, productivity and user satisfaction, as well as detailed descriptions of usability problems.

There is usability testing

Then a working group is assembled, which includes a product manager, a project manager, a technologist, and sometimes a representative of technical services involved in implementation. At the meeting, a plan of changes is drawn up and the next iteration begins.

It is clear that the process of iterative design and research is quite laborious, now our resources are not enough for all the products of the company.

Both technical interfaces (WAP, USS, WEB, IVR, and so on) and advertising layouts, such as promotional flyers, from which users learn about how to use services, and sites where users search for information and decide on purchase. Even television and outdoor advertising in its own way is the user interface, and the study of its perception can greatly increase its effectiveness.

What research brings companies?

Now the average level of convenience of services brought to the market without involving usability works in them is at the level of 45-60%. This means that no more than this share of service users are able to use them independently, without outside help and support. The rest are forced to experience difficulties, call technical support, make additional, stressing our network attempts to get the desired service, ask questions to relatives and friends, or even refuse from too complex services. All this is measured in money. At the same time, in the service, where the factor of convenience is observed, the proportion of successful users increases to 75-85%.

Illustrative examples of successfully implemented usability-design are WAP-portal Beeline, Hi service, Mobile commerce services, updated navigation through the main Beeline voice portal 0611, as well as a number of services accompanying voice communication.