As a UX designer, you will likely be involved in conducting user research as it forms a critical part of the UX process. For any upcoming researcher, engaging with people in real-life contexts and settings may seem like a daunting task. This post is aimed to alleviate some of those very concerns. This post thus serves a dual purpose. One is to have a documentation of my own experience of doing user’s interviews with the hope that it helps me build on the experience to make improvements for future opportunities of user interviews. The other is to share my experience so that it may work as a guide to others in a similar position. I hope this post helps other designers with improving their process. As most other posts on user interviews usually speak about the do’s and don’t of user interview sessions, I wanted to focus on the complete process and highlight observations and insights derived from my personal experience.
So let’s talk about user interviews. Defined in simple terms, user interviews are a research method where questions are asked to get insights about a certain topic, product, app or a website from the users.
User interviews help to understand:
How users perceive a product/website or the topic you introduce during the interview.
Reveal user perspectives and experiences which might challenge your own standpoints as a designer.
Help discover the issues and pain points the user faces while accomplishing a task; using the website, app or a product.
This method can also be used as a guiding force for the development of newer products: by researching the problems and needs of potential users and then developing solutions and products based on those findings.
Let’s start with the process first.
Preparing scripts for the interview
Before starting the user interviews, the first step is to prepare a script for the users. Script lays the foundation for the user interviews. It is the preparation that a researcher does before confronting the users for the interview. The script consists of questions that can extract vital information and insights which will help in going forward in the UX process.
To understand scriptwriting for user interviews, the following sections will be covered:
a) Constructing tasks/questions for the users
b) Structuring the tasks/questions
c) Getting questions reviewed by your colleagues
d) Preparing the introduction of the topic and reasons
for conducting the research
a) Constructing tasks/questions for the users
What are the key steps and things to keep in mind while constructing tasks or questions for the users? lets have a look:
Step 1: Setting up goals
It is important to understand what kind of information you are looking to extract from the user. In order to do this, it is good to set up a meeting with the client to understand the goals of the user when they use their product. An understanding of user goals helps us construct questions or tasks that the users will undertake when they interact with the website. Once the goals were set, the next step was to start framing questions or tasks for the user.
Step 2: Preparing the questions
Though the questions or tasks are prepared by the researcher can change depending on the situation during the interview, these set of questions/tasks act as a guide to the interviewer. The interviewer can scan through the questions while taking the interview.
These questions can also be helpful if you are working as part of a bigger team with multiple interviewers to avoid any misunderstanding.
We prepared the tasks/questions with cognisance of the goals that the user is trying to achieve on the site. While preparing the questions, the following points were considered:
Keeping the questions/tasks open-ended
We avoided asking any leading questions to the users. It’s crucial to keep this in mind as it could corrupt the data gathered. A leading question can influence the user to give a certain kind of response which will not provide useful insights. For example, ‘Why do you like this website?’ This question already declared the website likeable. A better way could be ‘Why do you use this website?’ This question is more open-ended.
Construct non-leading questions
Another vital point to consider while constructing the questions was that the questions should try to avoid using the exact terminology as used on the website. For example, asking the user ‘how would you search for a t-shirt on the website’ leads the user to take ‘search’ action when there could be other possible ways in which user can try to look for a t-shirt. A better way of asking this could be ‘How would you look for a t-shirt on this site?’ In this question, the user is open to looking for multiple ways in which the task can be accomplished.
Preparing follow-up questions:
Presuming expected responses to questions/tasks can help the researcher plan follow-up questions/tasks. In our case, when questions were prepared, all different kinds of responses and scenarios were assumed in order to prepare the next set of questions.
If a user is asked ‘how will you look for baking products on this site?’ on a grocery site, then assuming that at this point the user might look into the search or filter options, the researcher can plan the question ‘What will be your next step once you have found the products’ or ‘how would you narrow down your research to find products of a particular brand or with x features.
Another possible assumption could be that the user gets stuck on a question and their answer is ‘I don’t know, I only come to look for Utensils’. Then the question can be rephrased as ‘Can you tell me how would you look for those?’ Encourage them to think out loud to you to understand what they are trying to achieve.
There can be times when the user might also interact or undertake this task in a way that the interviewer did not expect. It is important to listen to them carefully so that follow up questions related to their interactions with the site can be asked.
Once these set of questions are ready, the next step is to go through the questions and structure them.
b) Structuring the questions
It is important to structure the questions in an orderly manner as it becomes easy for the researcher to scan through questions based on their relevance to a particular user. When you are asking questions you are involved in multitasking; you are listening to the user, you are looking at the time and trying to manage it in a way that you cover all the questions. At the same time, you are also observing and constructing follow-up questions that are related to this task.
For example, if you have the following info:
- Types of users — new, existing user or if its an interview where the task changes on the basis of gender or location etc.
- Based on tasks — Structure the tasks in the order that they take place when the user interacts with the product or an app a website or any other UX problem.
Then the structure more or less should look like:
- New User:
- Existing User:
Once the structure is ready, the script can be given to your colleagues for feedback
c) Getting the questions reviewed by your colleagues
Once the list of questions/tasks is prepared, your colleagues can be asked to review the same. This exercise particularly helped me in understanding if the users will be able to comprehend the questions as they were intended.
Once you have prepared the questions, the next thing is to prepare the introduction of the script and how you would end the interview:
d) Preparing introduction that establishes the context of the user interview with the user
The introduction should make the users comfortable so that they can answer your questions in a comfortable and open manner. Users tend to feel anxious before the interview and as an interviewer, the aim should be to curb their anxiety and make them comfortable. In order to do so, introduce yourself, who you are and why you are conducting the interview and how you will be using this interview.
A thank you note in the end is a must-have to thank the users for taking out their time, plus it helps make sure that the session does not end abruptly.
Preparing before the sessions
Practice makes perfect and having preparatory sessions can help you as a researcher gain confidence before the actual interviews. In our case, a mock user interview session was conducted with the help of colleagues, where one of the colleagues pretended to be the user and another took notes. This can also help reveal key gaps and technical issues that you can still fix before taking on the final interviews. It was particularly helpful for us as we were doing online interviews and user testing. Some other benefits of doing preparatory sessions are listed below:
Revealing if there are any unknown technical issues about the tool that is used to take interviews (GotoMeeting in our case)
Since the user interviews were taken over a call, it was important to thoroughly understand the application used for the call to avoid any panicky situations for the researchers later.
This also helped reveal that while talking to the users, the researcher should maintain pauses. Do not rush the script to fill the silences and wait for the users to respond.
The narration of the interviewer can also be observed by listening to the recording from the practice sessions. Reading the scripts out loud a few times improves the narration as the script gets memorised, helping the researcher remember what comes next.
It makes the communication smoother, as the researcher is free to focus on the answers of the interviewees and not fiddling around with the scripted questions. As the focus is on the answers, it’s easier to construct follow-up questions.
This article by Jane Austin was shared by our manager, it was particularly helpful in improving my communication for the user interviews.
Ethical issues in user research
For any research involving human participants, ethical issues must be taken into consideration. While doing any user interviews or user testing, it is vital that issues of consent, anonymity and confidentiality are kept in mind throughout the research process.
In our case, a consent form was sent prior to the interview to make users aware of how this information was going to be used and ask their consent on it. During the interview, we asked for consent on record as well to make sure the users were comfortable sharing the information. We also made sure we provide the users test accounts if they were not willing to share their own account for usability testing.
Learning and take away from the actual interviews
First interviews are usually not the best interviews.
Your very first interviews might not match your expectations but they do help reveal where your questionnaire is lacking and how the questions/tasks are actually perceived by the users. This is also the first time you are learning to construct follow-up questions which may seem difficult at first but gradually with more interviews, you will develop a greater understanding and may already know what to ask. So don't panic if the first interview is not so great as the process becomes easier as you go from one interview to another.
There can also be difficult users, they might think you have been asking obvious questions and not answer as per your expectations. Keep yourself neutral. Being aggressive can put the user into an anxious state and being too friendly may make them feel they have to meet your expectations and they might end up being too polite to share their problems or critique the product/website.
User interviews can be a really useful way to extract vital insights about user needs, expectations and experience of a product or service under investigation. But to get these insights one has to carefully craft the questions, ask these questions in an empathetic manner and listen to the users carefully. The process of user interviews does not end with the interviews itself as the data has to be transcribed and analysed in greater detail, but that’s a topic for another post.
But while doing the actual interviews, It is also very important to follow ethical practices and ensure confidentiality, anonymity and production of users data.
Hope this detailed documentation of my experience of doing user interviews proves useful in your own journey as UX researchers.