Doing user interviews is not easy, be careful to be led astray by users

User interviews can help the team to understand user needs more clearly, obtain relatively real user feedback, and then help the product to carry out subsequent iterative optimization. However, many product newcomers communicate with users, and often the so-called needs and pain points that they finally get are either fake or exaggerated, confusing priorities. In this article, the author summarizes some techniques to avoid being led astray by users. Let’s take a look.

It is said that when product managers and interaction designers participate in user research, in addition to watching what users do, they also have to pay attention to what they say, because users’ thinking and content ideas cannot be seen.

But it is very easy to let users speak, but it is very difficult to complete a valuable interview, because what the user says is always his personal “subjective” and cannot be used as objective material for reference.

Therefore, many product newcomers are very enthusiastic to participate in user interviews and communicate with users, but often, the so-called needs and pain points obtained in the end are either fake, or exaggerated, confusing priorities.

The author has many single-person and group interviews and research experience, and often communicates with online users. I hope that I can share this article to help you avoid being led astray by users.

1. Classic user interviews

01 User Interview Definition

User interviews originate from ethnographic interviews, which refer to the process of using question-and-communication methods to understand the process of user experience and obtain the feelings and thoughts of the interviewees.

Ethnographic Interviews: Ethnography means the systematic and in-depth study of human culture. Ethnographers spend years living in the cultures they study, trying to understand the behavior and social benefits of the culture as a whole. Ethnographic interviews borrow the essence and apply them at the micro level to understand human interactions and habits in relation to individual products.

The form of the interview is a conversation, so the topic of the conversation, that is, the question is very important. In the early stage of the interview, it is very important to design the interview questions well. Generally speaking, the questions include open-ended and closed-ended.

Open-ended questions are designed to encourage respondents to answer questions in detail, leading to more details on a topic to obtain more information. Closed-ended questions encourage short answers, and a yes or no answer is generally expected.

Investigators need to record key content when communicating with users. The interview content includes the product use process, feelings, etc. After the collected content is screened and organized, strong data can be formed to support product design.

02 Types of interviews

User interviewing is a huge professional field, and naturally it has its different categories and directions. We can simply divide user research methodology into individual interviews and group interviews according to the number of respondents.

1. Personal interview

As the name suggests, an interviewee conducts one-to-one or many-to-one interviews. In fact, individual interviews can be further subdivided into structured, unstructured, and semi-structured interviews, named according to how much control the researcher exerts over the interview according to pre-specified questions .

In addition to regular user interviews, you can also conduct interviews with domain experts, also known as expert interviews.

2. Team interview

The most classic is the focus group interview, and many of the current multi-person interviews are variants of it.

The problem of interview bias

Because professional and in-depth interviews require a special interview team, and for small and medium-sized enterprises, it is impossible and will not invest in doing things.

Then there will be the following problem: the interview becomes an immersive chat due to lack of professionalism.

01 Lack of professionalism in preliminary preparation

Many products or testers, when they hear what their superiors say, “find a few users to ask their needs” and “find users to ask questions about the current product use”, they often “ask” questions with a single mindset. .

For example, when you are working on learning software, you will ask questions like “What books do you mainly use for your studies?” “Do you like to use the lasso tool when taking notes on electronic devices?”

These are either big and wide, or small and narrow questions are not suitable, because these kinds of questions will make the content of the user’s chat become ineffective and worthless, and the details are too small to let go.

02 Unable to control the live conversation situation

When dealing with unstructured topics, researchers and interviewees often fall into endless topic derivation.

You can imagine some brainstorming meetings that the product team sometimes conducts. In the early stage, they may still focus on some topics. Later, because of unlimited requirements and unconsciousness, they will chat more and more, and finally find that they have not talked about anything.

For example, “What do you expect from the new smart TV operation in the future?”

Many users will talk about gesture recognition from the remote control, and from gesture recognition to arvr and the Metaverse, and they will be gone after half an hour of chat.

One-person interviews may be stopped in time by one party, and they can still be restrained.

Once it is a focus group interview, the pressure on the moderator will be even greater, because most users sometimes revise their speech due to the stimulation of the opinion leader’s speech, and will add subjective embellishment.

At this time, either the situation is a quarrel mode or a one-sided mode, the moderator has to constantly need to guide and promote the discussion, which is a very test of the moderator’s ability to control the field. (Better find a rational “social bullshit” host)

03 Users talk to each other

Nielsen: Opinions based on their own experience, the credibility is very low.

Users are different from experts, because they are invited to come, and they often come for rewards. Naturally speaking, many times they will be weighed.

For example, ask him “How do you feel after using this product from our company?”

The user will subconsciously answer “Very good, it will help me a lot in any scenario, much better than what”

Most of what is logged at this point is invalid content.

At this time, it is necessary to dig deep, or dig holes for users, to judge the inconsistent logic before and after.

3. Solutions to avoid going astray

01 Change the method at the source

The traditional single classical method needs to be changed. The current popular one is to use the Master/Apprentice Model, which is also called the situational research method in some places. This method is a combination of ethnographic interviews and immersive observation.

That is, the user is regarded as the master, and the interviewer is the apprentice. The apprentice observes the master and asks questions related to the user. The so-called master-disciple relationship appears naturally in the interview along with the content of the conversation, rather than being prepared in advance.

The biggest feature of this method is that the user intends to teach his own experience to the interviewer, and once he is ready to teach his own experience to others, he will not only talk about the conclusion, but will explain his own experience in order from start to finish. experience.


The interviewer (apprentice) “worships” under the door of the user (master)

Users demonstrate their experience while explaining

Interviewers ask questions when they encounter something they don’t understand

After listening to the complete experience, the interviewer retells the content of his understanding to the user to check whether there is any misunderstanding

Review, sort out the process, and let users evaluate

end thank you

02 Choose the right people when recruiting

Because the contents of the interviews are all spoken by people, who speaks is the key.

Generally speaking, there are several ways to recruit effective users.

One is the company’s resources, which can be filtered in the user library according to the corresponding tags, such as activity, product usage time, etc.

The second is the active users on the open platform, who often express their opinions on the corresponding type of products, but such users need to understand them before they can judge and use them, because some of them are critical kols.

The third is to find colleagues and friends who they are familiar with. This kind of user relationship will be higher, and future retrospective interviews and follow-up will be more convenient.

Six Dimension Theory: Looking for a friend of a friend, you can meet everyone in the world by looking for six times.

03 Set topics and questions in advance

In many cases the reason for confusion in interviews is mainly lack of planning.

Be sure to do a good job of the corresponding interview structure in the early stage , such as the difficulty distribution of questions, the proportion of open-ended and closed-ended questions.

Secondly, the problem description must be in a language that the user can understand . Don’t use professional terms to let the user know a little bit, so as to highlight the professionalism.

The last thing is to find a colleague to conduct a mock interview , make an advance, remove some irrelevant or time-wasting questions, try to control the rhythm and time of each interview, and don’t get entangled in ineffective links.

Fourth, the end

The most appropriate interview method depends on the purpose of the interview, the problem to be solved, and the stage in the life cycle.

The classic interview methodology is not intended to be used in the same way, nor is it to be used for reference , but to use its concepts in actual interview work, that is, qualitative and quantitative application modes, and a deep grasp of user emotional points.

In addition, don’t conduct interviews for the sake of interviews, and don’t turn them into brainstorming. Understand the results that each interview needs to get, and then reverse the process and content of the interviews, which will be more efficient.


This article was originally published by everyone is a product manager, and reprinting is prohibited without permission.

The title image is from Unsplash and is based on the CC0 protocol.