How to Get the Most Out of User Interviews / by Gavin Lau


  • Good for identifying attitudes, beliefs, and desires
  • Not good for specific instances, such as memory recall or speculation
  • Usually 20–30 minutes in length, but may vary with the complexity of the project
  • Best done in the early phases of a project, aids in persona development
  • Less likely to encounter the bandwagon effect than in focus groups, and therefore receive more accurate responses


Preparing for the interview

First, identify your research goals. What do you need to know about your users? How will that knowledge improve your product and inform your design process? Spend some time collaborating with your team to identify 2–3 clear goals. This will help keep your research focused and efficient.

Next, begin developing an interview questionnaire. It’s best to use open-ended questions, meaning no questions that can be answered with a simple “yes” or “no.” An open-ended question allows the interviewee to give the most accurate response because they aren’t being primed for any particular type of answer.

Avoid questions that frame the interviewee’s mind around a particular answer (called “leading” questions). For example, a question like “How frustrated were you with the voting process during the last election cycle?” assumes, first of all, that the user was frustrated with the experience, and second, influences the type of answer you’ll receive.

Better to phrase the question in a more neutral manner, like “Recall your last voting experience. What did you think or feel throughout the process?” Rephrased, this question makes no assumptions about the user’s experience at the polls and allows for a much broader range of answers.

It’s best to avoid questions that ask the user to recall or speculate on an experience because human memory is unreliable. But, if you must ask that type of question, try to be as specific as possible. Instead of “What do you feel when you have a bad experience at the polls?” ask something like “What did you feel last time you went to the polls?”

Throughout the session, be prepared to probe the interviewee for additional information as needed. Simply asking “why?” can often provide the clarity that the original answer lacked. However, if the respondent requests to skip a question, it’s best to oblige and maintain their trust.


Selecting participants

Your first step will be to select an interview method, as this will determine how you source your participants.

Interviews can be conducted a variety of ways: in-person, over the phone, video call, or instant message. In-person interviews provide the most behavioral data, as you can observe the participant first-hand. You’ll gain additional insights by observing body language and listening for verbal cues (tone, inflection, etc).

For example, if the interviewee pauses before answering a question, this likely means they are putting more thought into their response. Conversely, it could mean they don’t understand the question, and need further guidance.

How you source participants will depend on the type of project. If you’re working with a client, they can usually help you gather a list of potential contacts. If you’re not working with a client (or they’re unavailable to help), a recruiting agency can help gather participants for your study.

Are there certain people you know you don’t want to interview? To fix this, use a screener to eliminate participates who don’t meet your demographics. A screener is a set of pre-interview questions designed to eliminate respondents who don’t meet your criteria.

Once you have a list of participants, you’ll need to schedule a time (and place, if applicable) to conduct the interview. Provide as much clear, concrete information about when and where the interview will happen, and who will be involved. This will establish a base sense of trust before the interview has even begun.


Conducting the interview

Generally, the interview will follow a three-part structure to establish a base level of trust and ease the participant into the interview. Begin the session by introducing yourself, the company you work for, and the purpose of the interview. Make it clear what their information will be used for, and ask if they have any questions before you begin. This is key in establishing trust with the interviewee.

The first section includes your introduction and a few general questions about their day-to-day life. During this phase, try to develop a rapport with the interviewee, as this will put them at ease and increase their trust. Throughout the interview, it’s your job to keep the conversation on-topic and focused. Gently guide the interviewee back on topic if they stray, or go long on a question.

From here, you’ll work up to more detailed questions, and get to the heart of your interview. The second section dives deeper into the interviewee’s core frustrations and desires.

The third section returns to general questioning, and should gather more data about the interviewee’s hobbies and interests. Conclude the interview by asking for demographic information, but leave it open ended and allow the user to provide whatever information they feel comfortable with. Lastly, allow them to contribute any thoughts or feelings that they felt were not covered during the interview.


How to record and analyze your data

When possible, it’s best to record the conversation so you can refer back to it later. Whether or not you chose to record the conversation, I recommend taking notes on key observations and statements. Try to capture at least 2–3 direct quotes to include in the personas you’ll develop later on.

If you’ve taken good notes, the analyzing should be significantly easier. Using a third party tool, like OptimalWorkshop’s Reframer, can speed up the process. You and your team should look for patterns and themes that emerge across all the interviews. This will emerge through common answers to specific questions, key words, and similar demographic data. Then begin grouping interviews by commonalities, and each grouping will serve as the foundation for each of your persona types.


Pitfalls and things to avoid

Refer to the Neilsen Norman Group’s article on Interviewing Users for an excellent explanation of how user interviews can go wrong. If you’ve done well selecting the right participants and writing your questionnaire, you’ll naturally avoid most issues.

User interviews are a key step to gaining an in-depth understanding of your users and developing empathy for them. Your data will lay the foundation for the personas, which will become the common currency for discussing user needs through the project. For more reading on personas, I recommend this article from the Neilsen Norman Group.

What sources I referenced for this post, and articles for further recommended reading: