Effective User Interviews

Effective User Interviews

As a product manager, your objective is to create products that solve user pain. To solve the pain of your users, you need to understand them first - and user interviews are a fantastic way to get to know your user better.

One of the challenges with user interviews is knowing how to get started. After all, very few product managers are formally trained on how to interview users.

Therefore, in this guide, I’ll lay out clear steps to prepare for a user interview, to conduct a user interview, and to capture learnings from the user interview. This guide is structured the following way:

  • Before the Interview
    • Plan
    • Source
    • Screen
    • Schedule
  • During the Interview
    • Contextualize
    • Dive
    • Replicate
  • After the Interview
    • Retrospect
    • Document

Furthermore, I’ll provide an example as we run through the guide. I’ll pretend to conduct user research for Instacart.

Before the Interview

Plan

Determine what your objective is for conducting the user interview. After all, you shouldn’t ask users for their thoughts if you’re not sure what you’re looking for in the first place!

Are you trying to better understand a particular problem? Do you have a hypothesis in mind that you’re trying to prove or disprove? Is there a product decision that you’re trying to make?

Based on your objective, come up with questions that will help you achieve your objective. The kinds of questions you come up with will then drive how long your interview should be.

Note that most user interviews shouldn’t go past 60 minutes, since interviewees tend to run out of energy and focus by then.

In our example, let’s say that our goal is to get more users to try out Instacart for the first time.

I might structure my high-level questions like this then:

  • What are the different customer segments out there in the world?
  • How does each one shop for groceries currently?
  • How do they make decisions on how and where to shop for groceries?
  • What prevents them from using an on-demand grocery delivery in general?
  • Have they tried such a in the past, whether it’s ours or a competitor’s? If so, what were their thoughts? If not, why not?

Note how I have a particular objective: “get users to try out Instacart for the first time.” I broke down the objective sequentially.

To get users, I need to know the different kinds of users that exist, and what their behaviors are within each segment.

To have them try out Instacart for the first time, I need to know whether they’ve tried other grocery services before, or if they’ve never tried it before. I also need to know why they didn’t continue to use those services, or what’s stopping them from trying it out.

Once you have the plan for your interview, write up a rough outline. Print out copies for yourself so that you can refer to it as a guide.

The goal isn’t to set the interview guide in stone, however.

Much of the time, you’ll find that many new insights come from asking questions on-the-fly rather than sticking to your guide. Rather, having the questions in front of you will enable you to jump back on track when you need to.

Source

To conduct interviews, we need interviewees. I realize how dumb that sounds, but finding interviewees is honestly one of the hardest parts about user research.

We need to find willing participants. One great way to jumpstart your pool of interviewee candidates is to start with family, friends, and colleagues as initial interviewees, and to ask them to refer others to you.

As you expand your pool outward, you might find that you need to offer monetary incentives, especially if your particular user base is niche and if their time is extremely valuable.

For example, if you’re looking to ship a productivity app for , you probably need to incentivize to come take your survey. Generally speaking, you’ll want to consider your users’ compensation, then provide 1.5x that compensation for their time.

Here’s a concrete example. Say that you’re looking to speak with accountants for 30 minutes, and you know that accountants generally make $35/hour. They would normally make $35 x 0.5 = $17.50 in 30 minutes. You’ll want to offer 1.5x that, so $17.50 x 1.5 = $26.25. Round that up to $30 per interviewee.

As you look for participants, consider both unpaid and paid channels. Unpaid channels include word-of-mouth and postings, though note that you probably need to offer a reward to participants if you post online. Paid channels include social media, mailing lists, and newspaper ads.

In cases where you’re finding lots of difficulty in recruiting, considering hiring a third-party professional sourcer. Note that sourcers are generally quite expensive and only really helpful if your user base is particularly niche. One nice aspect about sourcing externally, however, is that they’ll also help you screen the participant list - which is what we’re going to dive into next.

Screen

Now that you have a pool of willing participants, you need to ensure that the users you interview are not biased. Biases come in two broad categories: beliefs and demographics. To account for these biases, you need to ensure that the sample you take is representative of the market that you want to tackle.

Jumping back to our Instacart example, let’s say I want to tackle the United States groceries market.

We shouldn’t exclusively limit ourselves to interviewing mothers. Other kinds of people shop for groceries as well, such as fathers, students, young professionals, and retirees.

Similarly, we shouldn’t exclusively limit ourselves to a particular geography unless that’s the specific geography we want to tackle.

Therefore, when we ask for people to join us for user interviews, we need to have some way to screen so that we don’t wind up with a bias.

A screener can be as easy as a Google form - just ask a couple of questions from the user. For example, my screener for the Instacart user interview might look like the following:

Thank you for your interest! Before we get started, we have a couple of quick questions for you. Once you’re done, we’ll reach out if we see a fit.

  • How often do you shop for groceries?
    • Never
    • Less than 1x per month
    • 1x per month
    • 1x per week
    • More than 1x per week
  • How old are you? 
  • How many people are in your household? 

Schedule

Once you’ve screened and qualified your interviewee candidates, you need to schedule them for interviews.

Use a calendar like Calendly to ensure that you can provide the most possible interview slots. Trust me, it’s a huge pain to coordinate back-and-forth with interviewees over email - it’s much easier if your calendar is open to your interviewees so that they can find times that work for them.

That being said, be sure to leave time between interviews so that you’re not rushed from one to another. You want to ensure that you have some amount of buffer time in between interviews, especially where you find a thread of inquiry that’s particularly insightful but requires additional time from what you had originally scheduled.

I recommend a 30 minute gap at minimum between interviews. Generally, you don’t want to do more than 3 or 4 interviews a day, since you’ll wind up with fatigue and won’t be able to perform well on subsequent ones.

Congratulations! Now you’ve got a strong set of interviewees to talk to. Next up, we’ll tackle how to conduct the interview itself.

During the Interview

Contextualize

When you begin the interview, provide context to your interviewee. Who are you, and what are you generally looking to learn about? Then, ask them to introduce themselves as well, in whatever way they’d like.

After all, your user is a human being. Your product is only a small part of her life. Therefore, you need to better understand who she is.

Of course, you always run the risk of being too broad. For example, it may not help to know that her lifelong dream is to become a firefighter if you’re trying to figure out a more effective note-taking tool for customer success managers.

Stay high level and keep your product out of the discussion for now. What is she trying to accomplish, and why is she trying to accomplish it? What’s blocking her from being more effective?

Let’s jump back to my example with Instacart. For their introduction, I want to learn more about what their background is: job, family, and life goals.

That helps me better understand their experiences when it comes to shopping for groceries.

Is it fun for them to shop in person? Or is it just a major pain? Do they have lots of free time, or none at all?

Keep the introduction to about 5 minutes. You don’t want to cut it short because you’ll lose context, but you also don’t want it to consume valuable time to find other insights.

Dive

Now that you have a sense of who your interviewee is, tackling the questions that you had in your guide.

Your most important objective is to learn how to think the way your interviewee thinks. You want to be able to faithfully represent them to your internal stakeholders, and that means being able to understand how your users make decisions.

As you progress through the interview, keep the following best practices in mind.

First, ask questions in an unbiased way. Instead of asking, “Why is it so painful for you to shop for groceries?”, ask “How do you feel about shopping for groceries?”

If you introduce the word “painful”, you bias your interviewee to interpret the experience as painful. Let them tell you whether it’s a pleasurable experience or a painful one.

Second, don’t ask questions that can be answered with either “yes” or “no.” For example, don’t ask “Are you the main person in your household who shops for groceries?” Instead, ask “Who usually shops for groceries in your household?”

Third, ask “why” to any answer that you get back. You may feel embarrassed, or you may feel the answer is complete - but many times, you’ll find that going one level deeper will yield new insights or ways of thinking.

Again, your goal is to understand your user all the way down to her thought process - the goal is not to mimic her on the surface.

Furthermore, when you do get a response, sometimes it pays to stay silent for an extra second. That moment of discomfort can trigger your interviewee to share more information that she may have felt wasn’t immediately relevant. It’s these kinds of spontaneous insights that often are the most valuable.

The hardest part about diving is getting the right depth on a particular topic. On the one hand, you want to keep pushing your interviewee to provide why. On the other hand, you also have other questions that need to get answered.

Don’t fret! Your interview is just one of many. Try to go deep every time - if there’s something missing, you can always ask a different interviewee for those insights.

Replicate

Close out the interview by running through hypothetical scenarios with your user, or by running through wireframes and prototypes. Now is the time to see whether their behavior is the same as their stated thought process!

Many times, interviewees will believe that they make decisions a particular way, but they actually behave in an entirely different way. By running through concrete scenarios, you’ll be able to see their thought process live.

When you see a contradiction in live behavior versus stated thought process, gently ask your interviewee how they made their decision, and why they made it in a different way than they said they would have.

Remember, your ultimate goal is to ensure that you can fully reproduce the way that your interviewee makes decisions in the real world!

Another fantastic exercise to run here is to ask your interviewee to select between competitors, or to evaluate different value propositions against one another.

For example, I might ask my interviewee to rank Instacart (groceries on-demand) vs. Blue Apron (subscription meals) vs. Uber Eats (meal delivery) vs. doing their own grocery shopping. By understanding what they like and what they don’t like about competitors and replacements, I can further hone the product to better satisfy that particular customer segment.

Once you’ve completed this exercise, thank your interviewee for their time and close out the interview.

You’ll know you’ve done a good job with your interview if your interviewee actively volunteers to send you additional feedback or thoughts through email. Take them up on the offer, though be sure to remind them that future interactions are unpaid.

After the Interview

Retrospect

Once you complete the user interview, reflect on the experience immediately. Use whatever downtime you have until the next interview to reflect on what you’ve learned.

Are there other areas of inquiry that you should going after instead? Are particular questions not working the way you expected them to?

Continue honing your interview guide until you get to the key insights that you’re looking for. Feel free to add questions, drop questions, or re-word questions!

At some point, you’ll find that new interviewees are repeating some of the insights you’ve heard from other interviewees. This is a sign that you’ve found most of what’s available in your area of focus, and that you’re going to getting diminishing returns through user interviews.

Remember that interviews are meant to gather qualitative insight, not quantitative measurement. Save the quantitative validation for later, through surveys or in-product metrics.

Document

Gather all of your notes and recordings in one centralized location. That way, you can send others a single repository of raw data, which helps with creating internal transparency.

Create a spreadsheet or table, and enter in attributes about each of the interviews that you’ve conducted so far.

For example, I’ve found it helpful to list out each interview on a row of its own, the name of the user, age, gender, location, household size, household income, and particular preferences that they mentioned in the call.

Then, provide a direct link to the raw interview itself. That will help you synthesize the data at a later point, and enable you to quickly navigate all of the interviews you’ve conducted thus far.

Once you’ve completed all of your user interviews, synthesize all of your notes across all interviews. Look for patterns between different kinds of users, and highlight shared pain points or shared mental models.

Summary

Divide the user interview into three stages: before, during, and after.

At each stage, your goal is to dive deep into the user’s way of thinking. You want to be able to bring them to life in front of your team, long after you’ve spoken to the user.

At the same time, you want to ensure that you have representativeness and that you can quickly analyze the insights.

By thoughtfully planning out the interview, you’ll maximize your chances for impactful and relevant learnings.

By diving deep during the interview, you’ll find nuances and new hypotheses that you hadn’t considered before.

By reflecting on the interview and documenting your core findings, you’ll enable your entire team to better represent the user, and build out a better product.

 


Have thoughts that you'd like to contribute around user interviews? Chat with other product leaders around the world in our PMHQ Community!

Join 28,000+ Product People and Get a Free Copy of The PM Handbook and our Weekly Product Reads Newsletter

Subscribe to get:

  1. A free copy of the PM Handbook (60-page handbook featuring in-depth interviews with product managers at Google, Facebook, Twitter, and more)
  2. Weekly Product Reads (curated newsletter of weekly top product reads)
Powered by ConvertKit

Published in On the Job, Product Management, Product Manager, Skills

Tagged in

2 Responses to “Effective User Interviews”

  1. […] remember that you can always pair quantitative metrics with qualitative feedback and […]

  2. […] into a platform product manager role, I experienced a minor regression in my ability to conduct qualitative user interviews, because I no longer relied as heavily on that skill […]