Getting the Most From User Interviews Part 1: Set Up and Conducting Interviews
Moderating tests and interviews can seem like a daunting challenge. This article, part 1 in a 3 part series, will help you keep in mind the dos and don’ts, as well as some general tips to set up and conduct user interviews to get the most out of them. This guide is adapted from the book Think Like A UX Researcher by David Travis and Philip Hodgson.
Some General Tips
The room
You get honest and genuine responses when participants are comfortable. Choose a room that is quiet and inviting. You may want to provide refreshments as well. After all, your participants are giving up their time to help you. Think of them as guests in your home.
The question set
When conducting interviews, it is important to understand what the business wants to learn from this exercise. Many times, these questions are made in a vacuum. Be sure to collaborate with your team. You may have noticed I didn’t say “script”. Scripts are rigid. It is important you treat interviews like conversations. Allow them to run their natural course, but make sure you are asking follow up questions to explore each topic.
The rule of two
The most important part of interviewing a user is listening. You can’t focus on the user’s comments if you are frantically trying to take notes. Every interview should have two people: a moderator and a note taker. The moderator is focused on actively listening to the participant. The note taker is there to, well, take notes. They are also there to make sure all of your questions you came up with get asked.
Understand the type of data you are collecting
In order to make use of both your time and the participant, understand the type of data you are collecting. Different types of data require different question sets. Is your data being collected to help you market your product (attitudinal) or to help you build your product (behavioral).
Do
Make the participant feel comfortable and build a rapport
In order to get the best possible data from the interview, the participant must feel comfortable. Reassure and have a positive disposition.
Be aware of sensitive data
Sensitive data is not limited to name, address, phone, etc.. It can be things we may not immediately think of as sensitive. Examples include chronic diseases, disabilities, and other information that can potentially identify a participant. When interviewing, keep this in mind and let the participant know they should only give information they are comfortable giving.
Transition to the interview (sign forms)
Legally, we need to get consent. This is a perfect transition to get these forms signed and begin the interview. Here is a privacy and consent form template you can use. REMEMBER: Always get approval from legal.
Be a reporter: ask probing questions
Surface questions only answer a portion of what we want to learn. Asking probing and follow up questions are necessary to get to root causes of problems, and identify true user needs, wants, and motivations.
Allow the conversation to progress naturally, but know when to get back on track.
Using probing and follow up questions are a great way to get a participant talking about root causes, but sometimes it can derail an interview. Always keep the value of the information you are receiving from the participant in mind, and know when to get the interview back on track.
Interpret responses and ask for clarification to get the right context.
Most times, other members of the team are not going to watch or listen to recordings of user interviews. Even if they do, some responses can be interpreted multiple ways. To clear up any misinterpretations, ask for clarification and restate their response as interpreted by you.
Don’t
Talk too much
Talking too much during an interview can be damaging in a couple ways. The most obvious is it wastes time. We have limited opportunities to talk to our users, talking too much can use up precious minutes. The second, the more we speak as interviewers, the more chance there is to introduce bias.
How to avoid this:
Be mindful of the value of what is being said. You want to make participants feel comfortable and build a rapport, but ideally, we should keep it to under 5 minutes.
Explain the design
A lot of new researchers fall in to this pitfall. Especially if they were involved in the design. You open the application or website, and start to walk people through how to use it. We are proud of our creation, and want participants to see the value we see. This is damaging to research, because not only have we introduced bias, but now we don’t know what the participant’s true path is. The test data is compromised.
How to avoid this:
Start by asking questions. When the application or website loads, start by asking “What is this and what is the purpose of this? You always want to ask more questions than give answers.
Answer questions
This is closely tied to explaining the design. We welcome questions from participants because questions indicate there is a problem. The participant is confused or stuck. When we answer these questions, we are “leading the witness” so to speak. Our data is compromised.
How to avoid this:
Ask probing questions like “How do you think you do that?” or “What do you think it does?” This allows for honest and genuine answers from the participant. It gives us better data.
Give your opinion
When we give our opinion on a design, especially one we helped create, we are introducing bias. We are also influencing the opinions and actions of the participants. We are not interviewing to express our thoughts on a design, we want to hear participants.
How to avoid this:
There are a couple of ways to avoid this. The first is asking redirecting questions. For example, “You said this about the tool. Can you explain that in more detail?” The second way to avoid this is to politely decline to answer. For example, “I would love to share my opinion, but I’m more interested in hearing about [fill in the blank].” Always bring the focus back to the participant.
Conclusion
User interviews can be an important source of information when you are building research. It is important that you keep in mind how you conduct interviews so you don’t compromise your data. Seek truthful, non-biased responses by being aware of how you ask your questions. Keep an eye out for part 2 of this series: analyzing the data.
