Three tips for better user interviews

One way we get to know our users is by interviewing them: implementing a systematic way to gather information. This helps us understand why humans think and act in certain ways. We’ve covered why knowing your users is essential to your product's success. But how do we talk to our users? The process isn’t simple but it is straightforward, once you learn the art of conducting a user interview.

The purpose of interviewing users is to get to know them. In order to ensure that we do this in a structured manner, we need to ask the right people the right questions in the right way. We have another article covering 5 short lessons for effective user interviews. Here, let’s dive deeply into three aspects of interviewing: Matching your research objective with the users you interview, the indispensable need for preparation, and building up your tolerance for silence.

Build up your tolerance for silence

Silence is a non-verbal example of an interview probe. Verbal probes allow you to get more information from your interviewee (“tell me more!” or “anything else?”) but non-verbal probes can be even more powerful.

We discussed the importance of silence in our other article on interviewing. Learning to be comfortable with silence is a lifelong journey. It still makes me squirm when the silence stretches out, no one speaking, and I have a nearly irresistible urge to fill the void with another question. Just as it gets unbearable, however, my interviewee has the tendency to say something remarkable, something they never would have said if I had powered on. Every time this happens, I’m reminded again how that momentary discomfort is worth it. How can you learn to embrace the silence?

Try practicing with your family or friends. At first, you can tell them what you’re doing, but later, try to do this spontaneously. This will help you get accustomed to that space.
When you’re in an interview, don’t force yourself to embrace the silences right away. Instead, make a note at points in the interview where you could have stretched out the silence, the moments where you could have given the interviewee more time to respond and offer more information. This is another reason why it’s valuable to record your interviews – you get the opportunity to listen to your own interviewing techniques.
When you’re ready to try it out, and you get to a moment of silence in an interview, try taking a slow, deep breath. Concentrate on your breath instead of the silence, and see what happens by the time you have completely exhaled.

Find the right participants

As we discussed in our previous guide, it’s important to start with a well-thought-out research question or objective. This objective will help define whether your research is generative (deeper understanding of a user group), evaluative (testing newly created concepts or solutions with users), or a combination of the two.

Next, we need to speak to the right people: these individuals can be current users of a product, or someone who fits the profile of a potential user. This is to ensure that you’re talking to the right users in order to find valid answers to your pre-defined questions. An example from my last role: Our key user group was bicycle couriers. We knew many things about them, like their average age, average tenure with our company, and how many shifts they usually worked per week. We knew what kind of bike they rode, how much gear they went through, and how many orders they could fill on average, per hour or per shift.

But we didn’t know plenty of other things – what were their main pain points on their courier shifts? How was their experience of entering a restaurant or ringing a customer’s doorbell? How did they troubleshoot moments of friction on their shift? The best – and only – way to find out this information was to interact directly with the couriers. Our research was generative in nature, as our knowledge was at an early stage of being built up. Our research objective was to get to know their overall user journey. Therefore, we could talk to a fairly general group of couriers to gain insights.

Later, let’s say we wanted to test a new app feature that was meant to enhance the courier’s experience of finding a customer address. Our earlier generative research revealed that a notable subset of couriers wanted audio-only directions so they didn’t have to rely on looking at their phone while out on the bicycle. This feature would provide audio directions to the customer’s home, but was only applicable for couriers who used earbuds or earphones to find the location. At this point, it wouldn’t make sense to talk to a general group of couriers – we needed to talk to couriers who fit this particular use case.

Prepare, prepare, prepare

I’ve conducted hundreds of interviews in my career and I still get nervous before I start a one-on-one conversation with an interview participant. To a certain extent, nerves are good: They mean you care about the task in front of you, and you intend to take it seriously. But too many nerves could throw you off balance and make the process not go as smoothly as it could. Here are a few practical tips on organization, which should ensure that your nerves serve only to fuel your excellence as an interviewer:

Whether your interview is more structured or leaning toward unstructured, always prepare for the content of the session. Preparation could mean writing a script in advance, with at least bullet points detailing what you want to know at the end of the interview. Not sure how to get started?

Always begin with a research question that you want answered (preferably one that will be answered after analysis of a series of interviews with relevant users).
Now that you know the question you want answered, list what is interesting or problematic about your topic. What don’t you know? Write down every question you can think of that is relevant to this topic, that in some way contributes to answering your research question.
Take all the questions you’ve come up with and arrange them in a logical order. Spend some time thinking about how this order will be perceived by the interviewee. For example, do you want your interviewee to walk you through a particular process? Then order your questions sequentially according to this process.
Take a critical look at the ordered questions. Filter out irrelevant questions and combine any similar ones.
Always be sure to pilot the interview before running your official sessions. This is ideally done with a member of the target user group. The pilot can be used to check timing, the logical order of questions, and get feedback from your test participant about the interview experience.

Author

Janelle Ward

Principal Consultant

Author

Continue reading

Build better product experiences with AI-first user research