The Interview Balancing Act: How to Build Rapport with Participants

In my years of conducting research interviews, I've received feedback ranging from, "you can't smile or be so friendly toward participants, it's not a conversation, you are not equals" to "it's amazing how comfortable you are, don't change a thing!"

This dissonance can be confusing, especially early in your career. This extra layer of "acting" is overwhelming when your head is already spinning with everything you need to remember during a session.

When a manager first told me to rein it in, I went from smiling and nodding to silence and frowning (I can't help it, I have RBF). Instead of the friendly tone and personality, my participants faced a neutral robot. I didn't know how to interview in this way, so everything I said sounded rehearsed and inauthentic.

As a result, my interviews suffered; I couldn't get to the same depth and insights I had before, and I walked out of each session questioning my skills. At one point, it got so bad that I threw the whole thing out the window and went back to my old ways. My interviews got better, and I gained back the lost confidence.

However, I still had a lot to learn about hitting that perfect balance in sessions.

To be or not to be friendly

The big question I frequently get is, "How friendly should I be with participants?" There is a fear that if we are too friendly with participants, they won't give us valid or truthful information—and this can very well happen.

In fact, there is a bias called the friendliness bias, where the participant will agree with everything the researcher says and give positive feedback to whatever the researcher puts in front of them. The cause? A researcher being too friendly in the session.

This friendliness bias is similar to the more significant social desirability bias. We want people to like us and sometimes that means we will lie to get them to like us more. How does this manifest in research?

For example as a participant, I see a researcher nodding encouragingly as I give certain feedback, or I notice a researcher cheering me on as I move my mouse in certain directions. I may be confused or unsatisfied during the experience, but I don't want to reveal this information. It's mean, and I don't want to hurt people's feelings.

This example may sound exaggerated, but it's not. Try being a participant in a research study, and you will experience this effect.

Ultimately, this leads to how friendly we should (or shouldn't) be during our research sessions. The (infuriating) answer is that we should be nice to a certain extent. We aren't trying to make a friend, but we want this stranger to open up to us.

Strike the balance

Here are some techniques I use to strike a balance on the friendliness scale and get participants to open up.

Consider status

I believe I am equal to—or even a step-down from—the participant during research sessions. When I say step-down, I am talking about status and expertise. The participant is the expert, and I am the learner—they always know more than I do.

This technique manifests in a few different ways:

1. I always look to the participant for advice or expertise, reminding them that I am here to learn about their processes and experiences.
2. I keep an open mind and accept their feedback and experiences, even if it is "wrong." For instance, if someone explains a way they use a system or a product, and it is not how we intended, it doesn't matter.
3. I use language that makes sense to the user, eliminating technical and marketing jargon from my vocabulary unless they use those words. Then, I mirror the language that they use.

Meet their energy (or go one step above)

I love when I get on a call where the participant gesticulates and has high energy because I meet that energy head-on. I aim to match the energy they bring to the session for every participant.

I do this to the extent that many of my colleagues joke about me having many researcher personalities. I copy their speech cadence, gestures, and overall posture. I try to imitate them as much as possible, allowing me to meet them where they are and causing them to open up.

The only time I don't directly match a participant is when they bring no energy. In this case, I will go one step above (minimal energy) to pull information out of the participant. Sometimes, I will even crack a joke or attempt to relate to something they are saying with these participants.

Build a bond or a relatable moment

When I started, I used to jump right into research questions: "what do you do?" or "how do you feel about travelling?" I didn't warm up with any participants. By leaving this crucial part of the interview behind, my sessions were good but never great.

One of my mentors urged me to use the first five to seven minutes of a session to talk to the participant. I tried it once and never looked back.

A warm-up does three wonderful things:

1. Builds rapport with the participant
2. Gets the participant used to a conversational interview and open-ended questions
3. Makes the participant feel more comfortable with you

When I warm up with participants, I also try to find a relatable moment or a small bonding point. What is something that brings us together? During my warm-ups, I ask questions like:

• What do you like to do with your free time?
• What are some hobbies you like?
• What's something you've watched, read, seen recently that was interesting?

When participants respond to my questions, I don't just say, "okay, cool." Instead, I engage in a short conversation with them. For instance, one of my participants mentioned they recently started knitting.

I asked them how they got into it and what they were knitting. They said they were knitting a Pokémon called Charmander. I am a Pokémon nerd and have Charmander tattooed on my ankle. I mentioned this, and we bonded about Pokémon for a few minutes. The conversation flowed easily after that, and the participant really opened up to me.

Think about response tokens

You know the utterances "mmhmm," "uh-huh," "mm?" Those parts of our language are called response tokens. There are positive response tokens such as "yeah" and negative ones like "nuh-uh."

Although small and relatively unconscious, response tokens are traffic signals for conversation. They give each person an indication of what the other is thinking. For example, if you are talking to someone and they say, "huh," you infer they are confused, maybe disagree with you, or might be uninterested in the conversation.

Then think about if someone says, "yeah" and nods while you're speaking; this response and behavior eggs you on to continue speaking because you know they are having a positive response to your words.

Now, when it came to research sessions, I had a tendency of using way too many positive words and response tokens: "yes," "great," "yeah." Over the years, I've dialed this back a lot and now try to respond only with neutral response tokens, "okay," "mm-hm," "uh-huh."

Be a participant

The real meta of user research: become a research participant and research how researchers research! See how it feels when someone doesn't smile or respond, and then watch for social desirability bias.

When I started, I was a participant in many research sessions. It helped me understand how a participant would feel in certain situations. I then found patterns in specific actions/reactions.

For example, I wanted to shut down when researchers didn't smile at me. When they interrupted me, I wanted to stop talking. These sessions were valuable in coming up with my interviewing style.

Consider the method

Generally speaking, I am more friendly during generative research than usability testing. I smile and give more positive response tokens during generative research because it is a conversation.

I want the person to feel comfortable telling me about their life. Imagine telling stories of your experiences and struggles to someone completely straight-faced and unresponsive. It is not encouraging.

I control myself more during usability testing because I want the session to be more objective. I make sure to use neutral response tokens, nod and smile less, and generally keep emotions in check, especially when working with the prototype.

I don't want the participant to get any indication from me whether or not what they are doing is good. If you watched some of my usability tests, you could even see me struggle with this concept, even today.

Find your style

Finally, and most importantly, you need to find your rhythm and style. As I said before, I couldn't interview differently because I would sound robotic and inauthentic. This doesn't mean I can't switch up techniques, but my general style stays the same.

I always warm-up, try to find a bond, laugh when appropriate, and smile. I rarely read from a script, especially during generative research, because the conversation comes naturally.

How did I find my style? Practice. I practiced with friends and family, colleagues, and in the mirror. Eventually, I could practice with participants, tweaking my responses or reactions.

Practicing and finding what felt natural to me made the difference. When a session felt effortless, I noted what I did. And, when a session was tough, I reviewed it and tried to see where my style could have broken down.

Keep reviewing your interviews to find what feels comfortable and makes you sound and feel genuine in your questions.