How deeply do you understand your customers’ pain points and what influences their decisions? Qualitative marketing research, such as customer discovery interviews and ethnography, can yield valuable insights about customers’ needs. But startup advisor and HBS Senior Lecturer Julia Austin observes that often, entrepreneurs “do interviews wrong” without realizing it. If not executed carefully, customer interviews can send you in the wrong direction. What can you do to conduct improve customer interviews and ensure you’re getting relevant feedback? Serial entrepreneurs and product experts share four customer interviewing techniques that can help you uncover meaningful data.
We sat down with Julia Austin, Wendy Tsu, Partner of New Business Ventures for AlleyCorp, and Allison Mnookin, HBS Senior Lecturer. They provide practical tips for obtaining more meaningful feedback from interviews and gaining a deeper understanding of market needs. Austin and Mnookin review four interviewing techniques. Tsu shares concrete examples of how her team investigates customer-problem fit through an ongoing process of customer interviews. Both suggest ways to use customer interviews to validate your problem hypothesis and iterate to create products people want.
Customer Interviewing Techniques to Uncover Meaningful Data
- Create an interview guide before you begin
- Focus questions on behaviors and unmet needs, not intentions
- Listen more than you talk—do not pitch your idea
- Make time to debrief and reflect on your discoveries
Create an Interview Guide before You Begin
Many founders are eager to begin talking with potential customers. But getting the most value from interviews requires strategy, planning, and practice. Austin, who is an expert in product development and an advisor for dozens of early-stage startups including Drift, Lovepop, and Wistia, observed common mistakes most founders make when conducting customer interviews. As a leader of HBS Startup Bootcamp, she counsels founders to set aside time to create a somewhat detailed interview guideline as a first step in the customer discovery interview process.
Use an Interview Guide to:
- Describe what you hope to learn—what types of information you want to uncover
- Frame an introduction to your company, research goals & confidentiality
- Outline timing
- Identify and describe customer segment(s)
- Capture core questions for each segment
Start broad. What are your research objectives? Austin advises, “Write out what you want to learn from—and about—your customer.” Not the specific answer that you hope they’ll give, she clarifies. But what details would give your team a deeper understanding of the problem from users’ perspectives? Is there a process you’re unclear about? Who are your primary customer segments? Why?
Use your guide to draft answers to general questions customers may have: how will you describe yourself and your research goals? How will you protect their confidentiality?
After you’re clear on the type of knowledge you hope to gain, brainstorm a list of questions to ask that would reveal relevant answers. From that broader list, you will later generate three to five more specific questions around which you’ll build the interview.
Planning: Logistics of Interviews
While you don’t want to appear rushed or rigid, Austin recommends limiting interviews with target users to 30 minutes. If you attempt to schedule more time, “they may not have the time for you” and decline. Additionally, if you have too much time, the interviewee—or you—can grow tired or bored and any insights might become muddled.
Always have a plan. Go in knowing exactly what you’re going to ask and what you want to try to learn in the limited amount of time that you have.Julia Austin
While you should create a plan and endeavor to follow the script, Austin concedes that “there are times where you will interview people who will go to places unexpected, and that will be a big judgment call—do you want to let them go there?”
When to Deviate from Your Plan during the Interview
If an interviewee begins deviating from your prepared questions, ask yourself:
- Are they sharing something that gives you a new idea about what you might want to solve?
- Or are they distracted and taking you away from what you need to learn in the limited amount of time that you have with them?
Crafting a Customer Research Plan at AlleyCorp
As Partner of New Business Ventures for AlleyCorp, Tsu investigates new business ideas, focusing on early-stage investments and incubation methodologies for new ventures. Since 2017, she worked closely on the concept development and launch of CoEdition and Truebird while leading investing efforts in over a third of AlleyCorp’s portfolio.
When launching into customer research and trying to ascertain problem-customer fit, examines the problem top-down and bottom-up which can entail researching multiple customer segments. As her team crafts a research plan, they keep two core underlying questions at the forefront.
2 Core Questions to Inform Your Plan
- How can this process become more efficient?
- How can the process or product be more cost-effective?
Her team drafts a simple one-page document to share with potential customers that summarizes the problem. For instance, “We’ve just started looking into this problem and want your active thoughts. Is this a pain point that you’ve experienced?” That preliminary document helps guide the team to “engage in thoughtful interviews, our focus group interviews, and product discovery interviews.”
How Many Interviews Should You Do?
Focused Customer Segment
While no magic number exists, based on her experience, Austin learned that, if you have a focused set of questions and customer segment you’ll begin to see trends within five interviews. However, she notes, “if you have multiple questions or the demographics or other characteristics of whom you’re interviewing are quite different, then you need to do more interviews.”
Having a set of five interviews for a set of questions for a targeted audience is enough. You’ll quickly see trends within five interviews.Julia Austin
Multiple Customer Segments
When trying to identify new ventures for AlleyCorp to invest in, the team conducts a series of interviews or conversations with multiple customer segments. Tsu recommends starting with roughly ten interviews to learn more about pain points and see where commonalities exist.
The first set of conversations are discovery. We try to do ten-ish discovery interviews, where we’re merely there to solicit their pain.Wendy Tsu
Craft Questions That Address Key Behaviors & Unmet Needs
In framing your questions, think about the context of the problem and try to pinpoint key behavior. Asking “How? Where? How often? When? With whom?” can help. You also want to deepen your understanding of users’ unmet needs. Be sure to incorporate questions that explore how they experience the problem today.
- What solutions exist today?
- Have they explored new solutions?
- What are the shortcomings of those solutions?
With these broader discovery goals in mind, Austin recommends writing questions exactly how you’ll ask them. “Precise wording is very important,” she emphasizes because you have limited time and need to stay focused.
Inquire about Specific Past Behaviors—Not Future Intentions
Ask questions about specific past behaviors that aren’t based on intentions. For instance, answers to the question, “When did you last _____?” yields more reliable insights than “How likely are you to ____ ?” Write open-ended questions that can uncover emotions. And after listening to answers, follow-up with “why?” to try to uncover any emotion underlying the answer.
Avoid asking customers questions that require them to speculate how they’d react—”Tell me what you would do if this happened in your life?”—which Austin calls “crystal-balling.” Instead, ask them to reflect on and describe how they deal with the problem today.
People have the tendency to tell you what you want to hear, or they imagine how they might do something, but rarely do they do what they said they might do.Julia Austin
Tsu elaborates that an effective way to uncover real pain points and pinpoint the problem is to ask, “Tell me about the last time that you did X, Y, Z.” Listen carefully and observe them as they describe their actions in their own words, then ask, “what do you like about using X, Y, Z?” and “what do you not like about using X, Y, Z?” Always probe deeper by asking them to clarify “why?” frequently.
Examples of Specific, Time-Bound Questions
- When did you last do X, Y, Z?
- How often do you do X, Y, Z?
- Describe your last experience doing X, Y, Z.
- What did you do first? Why?
- How long did it take?
- What do you like about X, Y, Z? Why?
- What don’t you like? Why?
Asking precise questions requires customers to reflect on their actions. It helps you identify where along the journey they struggled. Use those answers as inputs in refining your problem hypothesis. Tsu observes, “At the end of the day if you can deliver a value prop that is better, cheaper, and/or faster—and ideally is operating on at least two of those spheres—customers will pay.”
Listen More Than You Talk—Do Not Pitch Your Idea
It may sound counterintuitive, but the important part of interviewing, Austin notes, is listening—”you really want to do is have your interviewee do most of the talking.”
When you’re interviewing you’re not trying to sell your product or even raise awareness about it.Julia Austin
The more that an interviewee feels like you’re describing a potential solution, Austin cautions, “the less likely they’re going to be to give you the information that you’re really looking for.”
Finding Objective People to Interview
It’s natural to start by interviewing people they know. But Austin cautions that can lead to bias because “people who know you will tend to give you answers because they like you or they want you to be successful.” To get more objective answers, she suggests finding “people who have no connection to you.” How do you do that?
Identifying potential customers to interview is often easier when you’re investigating “any kind of consumer product,” she acknowledges because you can “target folks just purely based on where they live or how they behave.” If your problem is tied to an industry, you need to get creative and identify places where you might reach users. For instance, if you were exploring a problem in the construction industry, you might opt to visit construction sites, learn where union meetings occur or attend meetups for people in subsets of the industry, such as architects, electricians, landscapers, etc.
Pose Open-Ended Questions
Most people know to avoid leading questions, such as “wouldn’t you agree that _____?” Creating open-ended questions helps you to avoid subtly influencing your customers. “The last thing that you want is to phrase a question in a way that assumes that customers have a problem that they haven’t actually said that they have,” Austin cautions.
You don’t want to give them any impression that they should be trying to please you by giving you the “right” answer.Julia Austin
Stay Focused on Discovering Pain Points
Echoing Austin’s perspective, Tsu cautions founders against introducing a potential solution. You don’t want to start “force-feeding them a thing that [you’re] not yet sure about.” Instead, use discovery interviews to establish trust and identify key decision-makers who will become customer segments. Plan to conduct a series of interviews with those customer segments to learn as much as you can about their pain points. Simply asking people to “describe your pain” can provide valuable information that will help you understand the problem better.
Use a Two-Person Team to Conduct Interviews
To guard against unintentional bias, customer interviews should not be conducted by just one person. At the same time, having more than two people conduct interviews “can be overwhelming for the person that you’re interviewing.”
Ideally, every interview should have a facilitator—the person who asks questions—and an observer—someone who takes notes and watches reactions—to ensure more complete and objective results. Whenever possible, founders should observe while a teammate uses the interview guide to ask questions. “People have the tendency to tell you what you want to hear” Austin notes, and when “we’re very passionate about a problem that we want to solve, we have a tendency to hear what we want to hear.” You’re less likely to unconsciously influence the person being interviewed or misinterpret results if you’re observing.
Pay Attention to Visual Cues
Before starting, obtain consent to record the interview and explain the confidentiality stance you developed for your guide. Recording an interview is optimal as it provides you with the full conversation that you can revisit at any time. Mobile phones provide an unobtrusive and easy way to capture conversations, but taping the interview doesn’t negate the need for a note-taker.
While recording, note-takers should carefully observe the person’s behavior. Body language can also provide cues on when to dig deeper. How do they react to questions? Does their body language provide any clues about how they might feel? For consistency, try to record notes in a similar format such as this downloadable template for customer interview notes.
Ending Customer Interviews
Ideally, you should tailor your questions to your customer segment. But all interviews should end with two key questions:
- What did I not ask that you think is important to this problem?
- Who else do you suggest I interview about this?
Make Time to Debrief & Reflect on Discoveries
Immediately after each round of customer interviews, the team conducting the interviews should debrief and share their observations. Some may be obvious. Tsu notes that if “every single one of our interviewers” identified a similar pain point, you know to focus your attention there.
Other more subtle reactions may be equally important. Austin notes that the observer “may have noticed that the person you were interviewing leaned back and was inquisitive, or shook their head and was really frustrated”—such observations need to be shared before they’re forgotten. Often, the person conducting the interview didn’t notice subtle behaviors but those, Austin notes, “can be a big tell in terms of where a real problem is.”
Generating Concepts from Discovery Interviews
After the initial discovery interviews, Tsu’s team creates another simple document conceptualizing a potential solution and shares it with a new set of decision-makers for feedback. Approaching discovery interviews as an ongoing process with new customer segments might yield potential solutions to the problem that you would not have imagined.
Comparing Pain Points of Different User Segments
Tsu shares a specific example of the system of academic publishing which different customer segments named as a big problem, but they experienced pain points in distinct ways. Librarians and university administrators described their greatest pain points in terms of cost. Academic journal subscriptions—most of which are controlled by one publishing company—can run up to $10 million annually. Whereas another core segment—academics who need to publish research in the journals—feel pain at the bureaucratic processes which prevent the timely sharing of information.
Unexpected Concepts & Iterations
As AlleyCorp investigated the problems customers face in academic publishing, the team uncovered additional problems and nuances to the problem that they hadn’t considered. Researchers’ needs during the Covid-19 pandemic provides one example. Tsu shares, “Different institutions are working on different research but there’s no way that an article’s going to be published” quickly under the current system. Learning that “there really isn’t a collaborative ecosystem—a curated Reddit for academics—where researchers can talk to each other about the latest research and figure out best practices” Tsu’s team discovered an untapped need in the pre-print space.
You end up doing a few rounds of these. Every time you get more and more insights and a point of view around what the concept should look like.Wendy Tsu
After several rounds of interviewing, you will have acquired a more complete understanding of the pain points your target user(s) face and have ideas on how to approach a solution to the problem. Austin encourages, “This is all part of the validation process and you will iterate on this several times before you get it right.” Want to review how to make sure you identified the right problem? Want tips and templates for creating personas? See “How Do You Turn Your Idea into a Product Users Want? Make Sure You Identified the Right Problem,” Part 1 of this series.
Summary of Interviewing Tips
- Do minimal talking. Listen carefully to responses and take notes as you listen.
- Ask short, open-ended questions. Resist the urge to pitch your product.
- Ask follow-up “why?” questions that explore root causes.
- Watch body language and expressions closely. Look for any visual clues as the person is speaking.
- Take notes (and/or record) the interview to capture quotes in the user’s words.
In “Customer Interviews: Tips, Do’s, and Don’ts,” Dr. Agostinho Almeida, Investment Manager with Promotora’s Venture Capital Unit, provides tips for improving customer interviews, including conducting in-person interviews that “allow you to make a connection and better read people’s facial expressions and body language.” In a helpful list of Dos and Don’ts, he emphasizes the importance of focusing on the problem—not your idea for a solution—and avoiding mistakes like asking interviewees to sign an NDA.
In The Mom: Test How to Talk to Customers & Learn If Your Business Is a Good Idea When Everyone Is Lying to You, Rob Fitzpatrick provides valuable tips and good examples of how to engage with customers to avoid getting biased answers.
In “Customer Interviews: Voice of the Customer and Jobs To Be Done” Nikki Elbaz shares three methods for defining needs and goals for customer interviews and describes the benefits and limitations of data gathered in customer interviews.
In “95 Ways to Find Customers For Discovery Research,” Jason Evanish provides a helpful list of resources for customer discovery.