User interviews are a research tool that can provide insights into users’ needs and beliefs, while also building empathy.  

Not all interviews qualify as research interviews. While journalistic interviews may have some similarities with user interviews, research interviews focus on nonjudgmentally and objectively gathering information about user needs, expectations, and interactions with a product or class of products. So, not every conversation with a user can be considered a research interview.

Unfortunately, even with the best intentions, it can be easy to bias or influence your participants’ responses. User interviews require a lot of attention to detail, and can fail for a number of different reasons, but if you focus on avoiding these common mistakes, you can ensure that your gathered data is objective, unbiased, and methodologically sound.

Common Interview-Facilitation Mistakes

1. Insufficient Rapport-Building

To some people, small talk might feel like an unnecessary chore or a waste of valuable interview time. However, rushing in and diving straight into a user interview without spending enough time building rapport will limit the quality (and quantity) of the data you’ll get from that interview.

During in-person interviews, one good way to build rapport without taking too much time is to personally pick up the participant from the waiting area (rather than delegating that task to a different person). Even if you don’t have a “waiting area,” or are doing interviews remotely, make sure to spend some time setting the tone and making your participant feel comfortable.

That said, it is also easy to take rapport building too far. Some interviewers — in their effort to build maximum rapport and seem as likable as possible — might start to share their own experiences to build solidarity and empathy with the participant. However, this approach can skew participants’ responses by shedding light on what the interviewer is interested in and may also cause participants to withdraw if there is an experience that they do not share with the interviewer.

To avoid this mistake:

  • Introduce yourself and ask participants about their day, beyond the scope of the interview. This type of interaction can provide valuable additional context about your customers and the lives they live outside of using our products and services.
  • Speak calmly and slowly and try not to talk more than the participant. Participants will often "mirror" the researcher, so speaking slowly will help them remain calm as well. It also will give participants time to think about their responses. Aim to have the participant speak for roughly 80% of the interview time.
  • Avoid the word “interview” in your interactions with the participant; refer to the interview as a “chat” instead, to make it seem less judgmental. The word “interview” is often associated with job interviews and can increase anxiety (hindering rapport as a result).

2. Not Enough "Probing" Questions

Probing, or the act of asking followup questions to gain specific, in-depth information, is effective for uncovering the motivations and rationale behind certain behaviors, attitudes, and perspectives. Probing questions like “How does that make you feel?”, “Why do you think that is?”, or “Can you tell me more about that?” can get participants to share more information or to clarify what they meant in a prior statement.

These types of questions can feel a bit awkward or intrusive if you are not accustomed to asking them often; however, not asking probing questions can limit the depth and specificity of your participants’ responses and can lead to misunderstandings and ambiguity in your research data. The researchers may find themselves trying to extrapolate meaning from user quotes (rather than relying on a direct explanation from the participant).

To avoid this mistake:

  • Plan optional followup questions for each interview question as you’re writing your interview guide.  Even if you don’t end up asking them, it’s good to have some prepared followup questions to fall back on. These should be specific to the issue you are investigating (within this scope) and range in levels of specificity from broader (good at not assuming an answer) to very precise (something you really want to know, but don’t want to bring up yourself unless absolutely needed).
  • Keep a generic probes bank at the top (or bottom) of your interview guide so that you can fall back on them when needed, rather than repetitively asking “Why?” Generic questions are good at not leading the user, but you do need that bank of them, because it quickly gets annoying for a respondent to repeatedly be asked the same generic question (e.g., “How does that make you feel?”)

3. Multitasking and Note Taking During the Interview

When you are the sole researcher on your team, it can be an especially difficult to  commit your full attention to what the user is saying. Still, this is important, not only because it allows you to best interpret your participant’s words and gestures, but also because it builds the right kind of relationship with the participant. Preparation is key: the more you already know what to do next, the less you have to think about what comes next and can “live in the moment” and pay full attention to the participant.

Diverting your eyes from the participant to check your phone, answer a text message, or checking your watch can signal disinterest; so, turn off unnecessary applications and silence your devices. That said, the most common culprit of disrupted interviews is taking notes while facilitating the interview. This practice is problematic not only because it makes it hard to keep up with what your participant is saying and the participant has to wait for you, but also because it takes your attention and eye contact away from the participant and erodes the rapport you’ve been struggling to build.

To make matters worse, taking notes only when participants talk about something related to your research question can signal to them that some information is more desirable or interesting than other. Participants will often try to “please” the interviewer or be as helpful as possible, so they might alter their behavior to provide what seems to be most interesting or desirable to the researcher (rather than what is most representative of their experience).

To avoid this mistake:

  • When possible, record the interview and get it transcribed afterward. A recording helps avoid misquotes and gives other researchers on your team access to the same “raw” data for later analysis. Be mindful that your data-management practices respect your participants’ privacy, so anonymize the recording as much as possible, and remind participants that what is shared is confidential. (Even if this information is stated in the emailed instructions, these disclaimers can sometimes seem like insincere “boilerplate” text, so be sure to also say it out loud).
  • If you cannot record, assign a designated notetaker. If, for any reason, it is not possible to record the interview, have an assigned notetaker, whose primary role is to take notes on the entire interview (not just areas of interest).

4. Allowing Observers to Influence the Interview

While interest in user research is good, an observer in the interview room can hinder the ability to build rapport, which, again, can limit how willing your participant will be to share intimate details about their experiences.

An interview is meant to be a friendly one-on-one conversation; as you add observers who can insert their own questions,  it can feel a lot more like an interrogation. While it is not necessarily a deal breaker to have an observer or two present during the interview, the less noticeable they are to your participant, the less awkward your interview will be.

To avoid this mistake:

  • Record the interviews (as stated above) and offer recordings to any interested stakeholders or team members afterward. This tactic keeps observers completely out of the room and ensures best conditions for building rapport. That said, it is also not ideal for helping stakeholders or clients build empathy and common ground, since they may watch only fragments of the interview (or not watch it at all).
  • Set expectations and ground rules for observers. The interview facilitator should be the only person asking questions of the participant. Since observers are not actively building rapport with participants (and sometimes, may not be trained facilitators), the questions they may ask might not be received positively or might yield only terse responses. Set expectations and ground rules in advance (perhaps in a separate email or calendar invitation), and hold a short debrief for observers after the session (or series of sessions) so that any additional research questions can be factored into future interview iterations. (Note: If you're running the interviews via video chat, it may help to choose a meeting tool which hides the list of attendees from participants.)
  • Limit the number of interview staff in the room to 3 people and keep them out of sight. — either through a two-way mirror or by placing them off to the side, out of the participant’s range of vision. If running interviews on video chat and you're unable to hide attendee lists, consider having observers remove their profile picture, or rename themselves to something less recognizable, like a random string of letters/numbers. Still, participants may be affected by the knowledge that they are being watched.

As a more general point, most stakeholders are only willing to devote limited time to observing user research. Even actual members of UX team may not have time to observe all research, especially if they are not researchers themselves. In both cases, it’s better to allocate most of these people’s limited time to observing behavioral sessions (such as usability testing) where they can watch users actually use the product.

5. Leading the Participant

The most compromising of all facilitator mistakes is inadvertently skewing participant behavior or priming. Some researchers may accidentally “show their cards” and reveal the intent of the research study too early, thus biasing the participant’s contributions.  

For example, if we told the participant, “We’re studying how much word-of-mouth recommendations impact real-estate purchases,” the participant may focus their responses on recent word-of-mouth recommendations. On the other hand, if we said, “We are studying how people go about their home-buying process,” the participant might reveal that they don’t use word-of-mouth recommendations at all, but spend a lot of time independently browsing online real-estate listings.

As seen earlier, body language can also prime participants. A slow head nod changing into a vigorous one might hint to a participant that their response suddenly got interesting. Similarly, how questions are phrased can change participants’ responses. For example, the question “How important is price when looking at lawn-maintenance equipment?” might yield a predictable response of “Yes, price is very important,” because it implies that price should be considered and most people would like to respond in a way that is socially desirable and acceptable. It also reminds people of price, even though some may not necessarily think of that immediately. A better question would be “What factors do you use when selecting lawn-maintenance equipment?”

To avoid this mistake:

  • When introducing yourself and the study, keep the study’s purpose relatively vague, to avoid priming users.
  • Keep questions as open-ended as possible and start topics broadly before narrowing down with followup questions.
  • Mind your body language and try to stay consistent throughout the entire interview. You don’t need to be stern or robotic, but ensure you stay relatively neutral, yet friendly and interested throughout.

Conclusion

Interviews are a method to learn about how your users view the world. To maximize the quality of your research,  spend the time to build rapport and focus on being an open-minded listener.