How to run insightful customer interviews

August 14, 2022 (2y ago)

“You can close more business in two months by becoming interested in other people than you can in two years by trying to get people interested in you.” — Dale Carnegie

Many founders shy away from customer interviews. One reason is they confuse customer interviews with user testing. User testing is in-person or over video and involves walking a customer through your software’s interface. You draw out their thoughts and reactions in order to improve the product experience. This is a very valuable form of research, but it’s not what we’re talking about here.

Here, we’re talking about conversations that center on learning. ‍

The "Golden Rule" of customer interviews

Imagine someone approaches you in a bar. You arrived early to meet a friend, but they’re stuck in traffic so you’re killing time with a happy hour deal. While you’re waiting, a stranger slides into the empty barstool next to you.

After asking for your name, he launches into a monologue. It’s been a tough year, you learn, but he’s rounding the corner because he finally landed a new job. Which is good for the kids because, man, have the prices for youth sports increased over the past three years…

You don’t catch the rest because you’ve tuned out. You’re watching the clock and swirling your glass, counting down the moments until you can excuse yourself.

If you’ve ever been in a similar situation, you know: the fastest way to lose someone’s attention is to focus the conversation on you.

This is why the Golden Rule of customer interviews is it’s about them and their context, not you and your product. Once you practice this, conversations—even interviews—with customers become much easier.

How do you pick who to interview?

Counterintuitively, when you’re deciding who to interview, you don’t start with who. You start with what. As in, what you don’t know: What is your biggest knowledge gap? Or, what product hypothesis are you trying to research?

From there, you want to pick someone who can help you bridge your knowledge gap or answer your hypothesis. For established companies, this is usually one of three types of customers: recent sign-ups, recent cancellations, and superstars (your best customers).

If you’re collecting engagement data via Mixpanel, Amplitude, or Intercom, these customers should be easy to identify. They’re the ones who just signed up, churned, or have the most success with your product. Here are the questions each type of customer can help you answer:

Recent sign-ups: Why did you sign up? What was the trigger? What outcome are you hoping to get to?

Recent cancellations: Why did you quit? Was your job done and you obtained the outcome you expected? Or did something else happen that caused the switch? Superstars:

However, very new companies rarely have customers. If your product doesn’t exist or doesn’t have any signups, your best bet is to interview one of these two groups:

Advocates for a competitor: People who use a competing product or advocate for a competitor. They can help you understand the value of a competing product. For example, what does it help them do? And how does it fit into their workflow? Keep in mind, this doesn’t have to be a direct competitor.

Advocates for your solution: People who are outspoken about the pain you’re addressing or the solution you’re providing. They can help you understand what’s painful about the problem they’re facing and what success looks like.

This raises the question of, “well, where do you find these people?” You can find the first group either directly through competitors or through a site such as that can provide a list of companies using certain software. A good place to find the second group is Twitter, Reddit, review sites, and forums.

What do you ask in customer interviews?

When you sit down to interview a customer, it’s important you know what to ask as well as why you’re asking it.

For what to ask, see the Interview Question cheat sheet.

To understand why you’re asking these questions, we need to circle back around to Jobs to be Done (JTBD). Remember, the basis of JTBD is that customers don’t buy products, they hire them to do a job. In interviews, you’re looking to get a much clearer picture of something called the JTBD journey.

Mapping the JTBD journey means identifying not only the customer’s context, but important factors within that context. Such as struggling moments, driving forces, potential solutions, motivations, and the desired outcome.

Imagine it’s 2011 and you’re trying to build a new video messaging company called Zoom. You don’t have any customers of your own to interview, so you pinpoint a few people who use WebEx (a competitor) regularly. Here’s the type of information you’re listening for when you conduct those interviews:

Struggling moments

Current or previous events and situations that caused your interviewee to change their status quo or seek a new solution.

Example: “Our company opened up a new office in New York and our Austin teamworks closely with them. All 20 of us need to meet every week over video to talk through what’s working and what isn’t.”

Driving forces

The emotions your interviewee wants to feel or wants to avoid feeling by changing their situation. These are a goldmine for identifying why people might sign up for and purchase your product.

Example: “It’s really frustrating to set up a big meeting and then not be able to hear the other person because the video is crap. It feels like I wasted an hour of my work day—sometimes more.”

Existing solutions

How your interviewee is currently solving the problems you identified.

Example: “I use WebEx 2-3 times a week to talk with coworkers.”

Alternative solutions

The alternative solutions your interviewee considered, before they settled on the current solution. You’re looking to understand their options and how they settled on the current approach. Also, what did their status quo look like beforehand?

Example: “Before I started doing video calls, I would direct dial someone or cc the whole team on an email. But the email chain always got long and complicated, and talking with just one person on the phone left everyone else out of the loop.”


What your interviewee is trying to accomplish by hiring a new solution. For example, do they know what they’re trying to achieve, but don’t know how to do it? Are they trying to avoid a behavior that currently frustrates them?

Example: “I need to be able to collaborate with the rest of my team so we can get our work done on time with less miscommunication.”

Desired outcomes

Your interviewee’s ultimate goal. This is how they believe their life will be better by solving the problem with a new solution.

Example: “I want an easy way to connect with my nationwide team, whether I’m in the office or around town, so we can move faster and stay ahead of competitors.”

What good is all this information?

Twenty or so years ago, Samsung was making cheap and imitative electronics. Manufacturers drove features and engineers built products around price and performance. Design was an afterthought, not a cornerstone.

But in the early 2000s, Samsung conducted some customer research around TVs. And they discovered that customers were less concerned about showcasing flashy features (e.g. big speakers) and more concerned about presenting a beautiful home. Turns out, TVs spend more time off than on, and customers think about them as pieces of furniture. This flipped conventional TV wisdom on its head and led Samsung to produce a new style of TVs specifically oriented to this customer feedback. Today, Samsung has held the largest share of the global TV market since 2006.

The companies who win are those who best understand and adapt to customers. Interviews are your richest source of customer insights.

How many interviews are enough?

It’s more important to create an interviewing habit than to check off an interviewing quota.

The goal is to get in a continuous discovery cadence (a key way you become customer-centric) rather than starting a huge project, doing 50 interviews in the first month, and then skipping interviews next 6 months. ‍

So, if you’re trying to test a specific hypothesis or answer a research question, consider doing a burst of 3 interviews this week and 3 next week. Repeat these cycles until you answer your question; until you stop learning, don’t stop interviewing. Cultivate deliberate practice and short cycles.