Posted on: 10 May 2023
Lotte van Gelderen
How to interview the right people, with the right questions
You’re running late for work. You step outside, and there is a blanket of snow falling from the sky, you haven’t shovelled your driveway, and the car is covered in snow. “WHY?” you ask the world. Why didn’t I look at the weather last night? Why didn’t I wake up just a smidge earlier to deal with this? Rest assured, everyone has at one point asked themselves these questions. It comes from humans’ intrinsic need to ask, whether we’re asking another person, our pets, or we’re screaming them into the ether. This need to understand, no matter how mundane the topic may be, is something we’ve carried around our entire lives.
We grow up needing to learn how to ask the right questions. We start with something plain and simple: why, answer, repeat. We bombard our parents or guardians with why until either the recipient gets frustrated and tells you to stop asking them questions, or until you feel you’ve understood. Thankfully, our skills develop as we age, and we learn how to ask just the right types of questions (they can still be why questions). We garner the information we need more quickly, and without pestering the recipient. We may also learn something we weren’t even looking for.
This very theme fuels our lives. It’s the reason our society progresses, and we (hopefully) see an upward trend in productivity and efficiency in how our world functions as we do this. “What can we do to make this process work more effectively?” Even interpersonal questions like “What am I benefiting from staying in a friendship that I am currently unhappy in?”
Applied to UX
This notion of asking the right questions to ourselves and others is the foundation of interviews. Investigative interviews, interviews on talk shows, and what we are going to focus on today, user interviews, are all examples of practices where the questions being asked are the primary reason we gather the appropriate information from the interviewee. With this preamble, let’s talk about “Mike’s Music” radio station, and why their ratings are on the floor.
Mike has a radio station, adeptly titled Mike’s Music. Mike wants to know why his station has been seeing a decline in ratings. Mike has most likely already started thinking about topics he wants to know more about. He may want to know what makes an audience member listen to his station, and inversely, why someone may not. For both scenarios, is this decision due to the hosts? The music being played? The interludes? Or is it simply the name of the station?
All these questions are trying to understand the participants’ reaction when they interact with the radio station and determine what qualities are changeable in favour of Mike and his station. If he wants to increase listeners’ satisfaction, he needs to understand what is going to make them satisfied. All of this can be done using an interviewer, and a couple of participants, starting with singling in on the question that needs to be answered:
Why are our ratings so low?
To start, Mike will determine the core question that he wants to answer. This question should be as broad as possible, while still being specific enough that the answer will be of use. “Why are my radio stations’ ratings so low?” This question is the key to creating the next set of questions, that will be what the participants of the interview will hear.
Who will give us the answers we need?
To answer the question of why Mike’s stations’ ratings are so low, we need to know who is going to give us the answers to the question. Mike should probably be interviewing people who:
- Have listened to the station, but for some reason do not do it as often, or not at all anymore
- Have a taste for radio music
- Enjoy (in general) the genre of music being played
Variables like this are important – they act as control variables for participants Mike wants to speak to. These variables are set in place so that we are not speaking to people who ‘hate music’, ‘hate all rock music’, or ‘have never heard Mike’s station, so do not have a tangible reason to dislike Mike’s station’. In the interviews, we want to identify reasons why someone does not listen to the station. It could be things like:
- A dislike of a host’s personality
- A dislike in the interludes between songs that is aimed to keep the station entertaining (ex. Guest stars, games or banter between hosts, etc)
- The editing of the transitions are not smooth
- The same song repeats too often within the same hour
To get to the root of someone’s potential dislike of the station, we need to create questions like this to get to the root of Mike’s music’s ratings.
The Interview process
Once a list of questions has been formulated, the interviewer is ready to start thinking about the interview process. The goal during the recruitment process is to select a variety of participants, to make sure that a large demographic of listeners can share their insights. Demographic criteria such as age, gender, and geography are common criteria to make sure the participant is as unbiased as possible.
The next step would be to distribute and receive a signed consent form from the participant, outlining that the participants’ information will not be shared with anyone other than the interviewer (and their immediate team), and that they consent to being recorded, either digitally, or by notetaking.
Sidebar: Structured vs Semi-Structured Interviews
There are two main types of interviews that a team can choose to conduct: structured and semi-structured. Structured interviews are built upon the idea that a participant must receive a core set of unbiased interview questions. We aim for answers that are not influenced by subjective wording, intonation, and stressors on certain words, when asking the questions to the participant. Questions like:
- Do you enjoy using this product, yes or no?
- How likely are you to recommend this product to a friend, very likely, somewhat likely, highly unlikely?
The questions are delivered the same across all conducted interviews, with as little deviation as possible. When asked for clarification, the interviewer must try their best to reword the question, without giving any more information than has already been given.
Semi-structured interviews look and sound a little different. Instead of a list of questions that the interviewer must cover, the interviewer is given what is called a “discussion guide”. The name explains the content: it is a tool to guide the interview flow, and to steer the conversation into a specific direction. The questions in these interviews are often augmented with ‘sub’ questions, or ‘probes’ where we ask for more in depth insights, clarifications or specific areas of interest.
The goal of a semi-structured interview is similar to a structured interview, where the interviewer is using the wording of their questions to gather information about a specific topic from the participant. However, in a semi-structured interview, the way it is done is more guided by the participants’ interaction with the questions, and therefore also at the interviewer’s discretion. If the interviewer feels like a question does not fit the flow of dialogue, they may change the wording to better suit the participant’s frame of reference.
Semantics of the Interview
It is important to note that while the interviewer may change their wording, and adapt to the participant, they are not asking leading questions (questions that steer the participant to answer the question in a certain way). This keeps the interview unbiased and impartial. If clarity is needed with the content of the question, the interviewer may refer back to previous questions for context clues, as well as slightly rephrase the questions so the participant understands.
Mike’s Music Interview
In the case of Mike’s Music, both interview styles have merit. Let us take Shannon the listener as an example: Shannon is put into a structured interview, and the interviewer will ask questions like:
- When do you mainly listen to Mike’s station, mornings, afternoon, or evening?
- On a scale of 1-10, how likely are you to listen to Mike’s music on your daily commute to work?
- Where do you usually listen to Mike’s music?
This way of interviewing is a form of ‘quantitative research’, meaning the data collected can be translated into numerical values. This could be helpful for Mike if he is interested in knowing the amount of people logging into his station at different times of the day, or how many people prefer his rock music over his country music selection.
The answers could be used by Mike if he simply wants to understand his audience better, and make changes such as airing time, what music he should play more of, and how far his station’s reach is (whether he has listeners in rural areas of his country for example). Structured interviews can gather objective data, and the data collected is usually easier to analyze, as the answers are usually more simplistic in nature. Furthermore, this way of interviewing is usually easy to conduct (due to the structured way of questioning), and therefore can collect large amounts of data.
The semi-structured interview looks quite different. This interview style will allow for interpretation of the questions by the participant, and may allow for the interviewer to really dive into what feelings are evoked from the radio station. This way of interviewing is used to gather descriptive data, and is coined ‘qualitative research’. Questions like:
- When you listen to the station, what influences whether you listen to the station for 15 minutes, or a whole hour?
- Why would you choose not to listen to the station during certain times of the day
- Are there aspects of the station that can be changed in favor of the station, if so, what are they?
- When you listened to (Mike’s music) station, what was your first impression? Why?
Mike’s final choice
Eventually, Mike’s research team decides to go with the semi-structured interview process. With all the data gathered, Mike will be able to extrapolate on not only the routine behaviors of his audience, but also understand why people exhibit these behaviors. He could use these findings to make educated changes to his station in areas where he may not have seen necessary: the transition of music to talk segment is too jarring, or the two hosts have very different personalities that clash.
While these are definitive statements to make, Mike can start to see patterns like these in the responses gathered from the semi-structured interviews. To elicit a positive response in his audience members, Mike first needs to know what the current response is, and he will be able to do just that with these interviews. He will be able to use the answers of the questions above to change the reaction people have to his pride and joy.
When someone (other than Mike) is thinking about which style of interview to use, or whether to use an interview at all, it is vital to understand exactly what one gets out of these research methods, and what one needs to know going into the research itself. As stated for the interview methods detailed above, the ability to have a 1:1 conversation (virtually or in person) with a client/audience member/user is necessary. For other research methods however, like a survey, this is not a necessity: one can format the questions similar to that of a structured interview, and distribute the survey en-mass.
This is advantageous for companies who have a large reach, but do not have the resources to speak to many people one by one and want a large sample size to base their conclusions on. Very often, interviews are followed by a survey, so the company or individual gets both the quality and quantity of each research method. The context in all these scenarios will truly be the distinguishing factor here: when to use one approach over another is based purely on what someone has included in the research, and what one wants to get out of their efforts.
Lotte van Gelderen
Lotte has always been interested in what drives people. Whether in the context of mental health or in UX, she has shifted her focus to what drives people to dislike and like in their daily interactions. Whether this is what candles they burn in their house, or which photoshopping tool they prefer using over all others – understanding people as a whole is what gives her energy. She aims to continue to try to understand what makes people tick and is happily progressing in this journey of discovery.