In interviews, people do not always tell the truth about what they are doing. The reason for this is not evil intent. This is the specific perception of habitual routine actions. No matter what kind of actions, the main thing is that they repeat constantly. Try on the fly to describe your typical work day. How detailed will the story be? Surely you miss a lot of details - it is difficult to remember the details, or the information seems to you not too valuable.

The problem is that in order to develop IT products, we need the user to describe in detail his daily activities. The created product should solve real problems and organically fit into the user's social and physical environment. How to get this description? One of the methods is contextual interviews (contextual inquiry).

The contextual interview is an attempt to combine observation and interview. We observe what the user is doing, ask him to comment on his actions, ask clarifying questions and with him describe in his daily routine as much as possible. So the method is implemented in practice. But in order to work effectively with him and not make critical mistakes, this information is not enough.

In this article, we will similarly consider the basic principles of working with contextual interviews at the preparatory stage and at the stage of field work and describe how to solve possible problems. Knowledge of these principles will help you avoid serious mistakes at first. The rest will come with experience.
')
The article will be useful:

• UX-researchers and managers of UX-departments;
• product managers;
• team leaders who work with professional software.

The following sources were used in the preparation of the article:

• Karen Holtzblatt, Hugh Beyer «Contextual Design. Evolved "( http://amzn.to/2pFlog0 );
• Jim Ross “Why Are Contextual Inquiries So Difficult?” ( Http://bit.ly/2pU5IVC ).

We recommend to get acquainted with these materials for deeper immersion in a subject.

Focus of research: where to look and what to ask

Let us once again designate the main problems that we will solve with the help of contextual interviews.

• Users can not describe in detail what they actually do. This is the problem of everyday routine practices.
• Users give too superficial description. They talk in detail about the problems, but not about the usual actions.
• Users forget about workarounds that they came up with to overcome difficulties when working with the product.

Our task is to get the most detailed description of user actions. We go to the user, see what he does, fix and discuss. Simple and clear.

And why are we going to watch? And what are we going to ask? The first desire is to see everything, to discuss everything. This is where the difficulties arise: the researcher is objectively unable to see and discuss everything in one session of the contextual interview. Given that the average session duration is one and a half to two hours. What is there to catch?

In fact, you do not need "everything" - so you need to understand exactly what interests you. After all, any study begins with the definition of the focus of the study. For contextual interviews this is especially important. The focus of the study shows what you need to pay attention to during field work. This does not mean that the researcher will ignore everything else, but he should focus on key topics. To identify the focus of the study, the project team must answer a few questions:

• What problem do we solve?
• Who are our users?
• What information do we need to work on the product?
• What actions and tasks of users interest us?
• Where and when does the user perform the actions of interest to us?

All these questions should be answered before the study begins. Reply and commit information.

The focus of the study determines not only the goals and priorities for data collection, but also the start of field work. The study is conducted where the user performs the actions of interest to us (for example, workplace, house, car, cafe, etc.). When planning sessions, you should make sure that the user will perform the necessary actions at the scheduled time. For example, we want to understand how a person generates a monthly report in professional software. It is advisable to catch him at the exact moment when he is working on this report. If the actions we are interested in are performed in several stages, you can break up the study into parts and come to the user several times. In more detail about the organization of field work we will talk further.

Fieldwork: organization and recommendations for data collection

The contextual interview method is based on four basic principles:

• partnership;
• focus;
• context;
• interpretation.

These principles determine the specifics of data collection and communication between the researcher, the user and the customer. Let's look at each principle in more detail.

Partnership

The most important of these principles is partnership. The contextual interview assumes that the researcher works in tandem with the user in order to compile a detailed description of his actions. The important point: the expert is a user, not a researcher. Necessary conditions for collaboration:

• the user correctly understands the purpose of the study;
• the user trusts the researcher.

The user should understand well: the purpose of the research is to study his daily activities in order to help him more effectively solve his tasks. We work to make his life easier. It is extremely important to explain that this is not a test of his competencies or the quality of the tasks, especially when it comes to studying the working day. And even more important for the Russian context: we promise that there will be no “witch hunt” after the research. All these points need to be discussed in advance with the user and explained to the customer.

This is necessary to establish a relationship of trust. Immediately before the session begins, the researcher must introduce himself and once again designate the research objectives. In addition, you can immediately discuss what tasks the user will have to perform. The researcher can write down what the user thinks about the goals of the project and how important these goals are for him.

The method of contextual interviews involves photo and video. For this you need to get user permission. The researcher must ensure the confidentiality of the data: the materials will be used only for research purposes. Usually, people do not mind recording if they understand the goals and guarantee anonymity.

Focus

We have already said how important it is to determine the focus of the study. Now we will focus on its practical application during field work. Contextual interviews are a flexible methodology. Before starting the study, we do not have a clear list of questions that need to be asked to the user. Given the specifics of the method, we cannot compose such questions in advance - they are directly dependent on the user's actions and comments.

Questions are formulated in real time when monitoring the user. This is where our research focus comes in handy, which can be presented in the form of key topics and a list of user tasks. This list will allow you to direct the user in accordance with the objectives of the study. But there can be no hard list of questions: the researcher is guided by the user's actions and context.

Context

Context is what we use contextual interviews for. It is for his understanding that we study the user in his natural environment, watching his actions. Our task is to eliminate the distinction between real practitioners and their retrospective description. The context includes both the physical and social environment of the user. What should a researcher pay attention to?

• User actions. What tasks does the user have? What does he do to solve them? It is important to find out how long these actions take, where and on which devices they are performed. Is the problem solved in one place or does the user finish part of the work from a mobile device? Is a break required to solve it? Maybe the problem is not solved in one working day? In which cases the user solves the problem quickly, and in which cases he is forced to fight the product, devising workarounds?
  Example

  The manager (user) needs to register a new client in the system. This can be done quickly by the number of his loyalty card in the affiliate network - enter 16 digits. But with this operation, the system often freezes, and the client starts to get angry. The user selects a workaround - passport registration. This method is longer, but safer. True, not all customers have a passport with them.
• Emotions user. What does the user like about his work? And what annoying? To understand the user's emotions, you need to follow his comments, gestures, facial expressions. It is also important to find out which aspects of the work are meaningful to the user. What is he proud of? What inspires him?
  Example

  The client department manager is satisfied when he can quickly solve a client's problem, choosing the right product for him. In addition, he gets a percentage of sales. The problem is that the system does not allow to find a product with incomplete characteristics. We have to look in paper catalogs, and it takes a lot of time. The manager is very upset because customers get annoyed and give him their complaints.
• Communication This is important for describing the user's social environment. Who does the user communicate with when performing their tasks? How often does he do it? Is it offline or online? What role do these people play in his daily activities? It is important to understand how everyday communication is organized and how it affects the user's actions.
  Example

  For the operation, the user needs to obtain approval from several employees. Some operations need to be carried out urgently, and employees do not always look at documents that are pending approval. The user has to personally approach each employee in order to ask for a quick confirmation of the operation. The problem is that some of them may be absent from the office and have to wait for him to return. All this delays the approval procedure.

For fixing the context, the researcher uses photo and video. You can fix:

• user;
• Desktop;
• environment (physical space);
• computer or other technical devices;
• screens with examples of the interface;
• documents used in the work;
• instructions.

Once again: all pictures and videos are made only with the consent of the user. Do not forget to tell you how to use these materials in the project.

Interpretation

A key feature of the contextual interview method is the joint interpretation. We remind you that the user in the context of the contextual interview is an expert. Together with him, the researcher carefully studies his actions and emotions. There is little simple observation for this. The user must comment on their actions, and the researcher must ask questions. During the session, the researcher shares his interpretations with the user in order to listen to his opinion and correct his understanding of his actions. In fact, the entire contextual interview session is such a cycle:

• the researcher monitors the user's actions;
• Researcher listens to user comments;
• the researcher asks clarifying questions;
• the user answers these questions;
• the researcher offers his own interpretation of user actions;
• user comments on this interpretation;
• the researcher makes adjustments to his interpretation.

This cycle is applicable both to individual user actions and to the session as a whole. It is important that the researcher is not afraid to seem stupid. You can, for example, tell the user: “Imagine that you are training an intern.” Of course, the right to ask stupid questions does not negate the need to thoroughly get acquainted with the subject area before the start of the project.

Although the main advantage of the contextual interview is the observation of what is happening “here and now”, the researcher may ask questions about the actions performed by the user outside the session. A good option would be to use artifacts, that is, referring to the context. For example, you can ask how the document was prepared, which lies on the user's table. Or why this sticker was pasted and what the information on it refers to. Such techniques help to better understand the daily activities of the user.

Problems when dealing with contextual interviews

Inspired? Contextual interviews allow you to collect a lot of valuable information about the user's daily activities. However, when working with this method, the researcher faces a number of problems. The main problems are described in detail in the article by Jim Ross, we briefly describe them in the context of the Russian context. (For a deeper immersion in the topic, we recommend that you familiarize yourself with the original - http://bit.ly/2pU5IVC .) All problems can be divided into two large groups:

• problems with the customer in organizing the study;
• problems with communication with users.

Let's start with the problems associated with the organization of the study.

• The customer does not understand what contextual interviews are. A Russian customer often does not know what a qualitative methodology is (more on this in our previous article http://bit.ly/2ptV2QS ). In this case, try to explain to the customer the essence of the methodology, focusing on the research results and their use by the project team. Separately, it is necessary to tell what is required to organize a contextual interview, which is conducted in a natural environment for the user.
• The customer does not understand that the contextual interview is a flexible method. He wants you to ask the user specific questions. As we have said, you cannot have a hard list of questions: all questions are formulated in real time during the observation of the user. Try to reorient the customer to develop the research focus and identify meaningful user tasks.
• The customer (or users) want to hold sessions in the conference room (dining room, "chat", cafe, remotely via Skype, etc.). Try to convince the customer that the research needs to be conducted in a natural environment for the user. This is the only way to get valuable contextual data. If you can not persuade, give up the contextual interviews and replace them, for example, the usual interviews with users. But do not forget to warn the customer that because of this, some of the valuable data about the user's real actions will be lost.
• The customer (or users) are trying to organize a group session. This moment needs to be discussed at a preparation stage: research assumes supervision over one person. Do not settle for group sessions - this reduces the quality of the data. The only permissible exception is the situation when two people perform one task. In this case, you can observe how they work together to understand how they interact with each other.
• Sessions organized are too short. Sometimes you miss a two-hour session to see all the tasks that interest you and the customer. Warn the customer that you may need to approach research planning more flexibly. For example, organize additional sessions. Or divide the tasks between the participants: it is better to get deep data on the same task, which is performed by different users, than the surface information on different tasks from one person.

Now consider the problems in communicating with users.

• Users during the session return to the interview format. For example, they are waiting for your questions, instead of commenting on their actions. To avoid such a problem, the researcher must tell the user in advance about the format of the study. During the session, the researcher can also remind the user what is being asked of him now. For example, let's do this task. What are you doing now? Why did you decide to act that way? Such reminders help the user to switch between “do and comment” and “answer questions” modes.
• Users do not show the solution to the end. Users do not always understand that you need to see the whole process of solving a problem - remind them about it. Another case is that the user has no real task right now. For example, a report has already been compiled. Here the output could be to create a fake report. It is important that the user actually does this operation, rather than performing it for show or telling how he could perform it. To make situations with missing tasks less frequent, carefully plan your research time. You can make a list in advance of the tasks that the user must perform during the session, and ask them to postpone this work until the study.
• Users say that the researcher does not understand anything (for example, they use highly specialized terms or professional jargon). Of course, the researcher must dive into the subject area before the start of the study, but still he will not become an expert. And it's not scary. Remember that we talked about silly questions? Do not be afraid to ask them. Remind users that you are not an expert in this field. Actively use the “mentor-trainee” model: let the user educate you by commenting on their actions.
• The contextual interview turns into a user complaint. This is a normal situation: it is easier for people to talk about what causes them pain. But you need to get a detailed description of the user's actions, not just his problems. Feel free to direct users. Promise that you will discuss all the pain points at the end of the session.

These are the main problems that arise in the process of working with contextual interviews. You probably noticed that many of them are solved at the organization stage. Problems with the customer can be avoided if you explain in detail to him the essence of the methodology and the specifics of the research organization. There will be less problems in communication with the user if you adhere to the principle of partnership: the user assists you in research, and you clearly formulate the purpose and objectives of your collaboration.

Conclusion

We reviewed the basic principles of working with contextual interviews. Understanding these principles will help you avoid serious mistakes at first. In addition, we analyzed the main problems that arise when communicating with the customer and users. The researcher must be prepared to solve these problems.

Contextual interviews provide an opportunity to collect a large array of high-quality data on the daily activities of users and thereby obtain the information necessary to develop high-quality IT products.

Processing the collected information was left out of the article, since the analysis of qualitative data — not only those obtained through contextual interviews — deserves a separate discussion. It is important to understand that after research it is necessary to allocate time for processing the collected data set. The lack of analysis time is one of the problems described in the article by Jim Ross. On average, one hour of a contextual interview session requires at least two hours to analyze the results. Be prepared for this. Without a thoughtful analysis of the data, proper registration of research findings and further work with them by the whole team, the method of contextual interview will be a waste of energy.