Conducting effective customer discovery interviews
How to get candid insights — even from your biggest cheerleaders
I started to learn how to conduct customer discovery interviews in the summer of 2020 through Stanford’s Hacking for Recovery program based on the school’s flagship entrepreneurial class Lean Launchpad. I was eventually able to enroll in the full course, and I have ever since iterated over 6 different business ideas.
I enjoyed the process so much that, after going to Costa Rica with my classmates for Spring Break, I prolonged my stay by 3 days as I wanted to conduct customer discovery interviews with the locals (I was working on an idea that could apply to the Costa Rican market). Here are some of the best practices I’ve learned along the way.
5 things to do
1. Ask open-ended questions
The ultimate goal of a customer discovery interview is to gather insights that are not biased, to educate you as you build on your idea. An open-ended question leaves room for your interviewee to offer you narratives and thereby generate insights you may not have expected.
For example, last year I was building an app enabling small business owners in Haiti (my home country) to do their bookkeeping digitally. Today it’s mostly done manually (pen and paper) if it’s done at all. Instead of asking a question like “do you record your transactions?”, it’s better to ask “how do you record your transactions?”.
The first question has a strict yes/no answer, whereas the second enables the interviewee to share their genuine experience.
2. Ask for real and recent examples
It’s easy for people to tell you how they feel about a situation instead of how they actually behave. Past (and recent) behavior is a good predictor of future conduct.
For example, in addition to asking the interviewees how they generally record their transactions, it’s effective to also ask them to narrate the last time they actually recorded a transaction. Asking for a real example will force them to be more concrete and give you valuable insights. In essence, ask for real and recent past examples to have a better understanding of how your interviewees actually behave.
3. Dare to ask for an engagement
In most situations, people will try to be nice to you and encourage you to keep working on your idea. As a result people often tell you what they think you want to hear, and it becomes difficult to understand how much they actually value what you’re building.
A way to test how much they actually value your offering is by asking them for a commitment. For example, earlier this year, I was working on an HR solution with some of my classmates. We interviewed a number of companies to understand their needs and ultimately pitched our idea. In most situations, the interviewees gave us positive feedback on our product — but when we asked if they would be open to doing a pilot, it was a resounding “no”. We thus stopped working on that product.
Other ways you can ask for a commitment is by asking the interviewee to pay for early access or having their friends sign up to be on a waitlist. If all they do is give you compliments, take it with a grain of salt.
Asking for a commitment can help you gauge how much your interviewees actually value your offering.
4. Ask for details and clarification often
When you’re conducting the interviews, don’t let anything slip. For example, back to our bookkeeping app idea, if an interviewee tells you that inputting their transactions in an app is difficult, don’t assume you understand what they mean. Ask them point-blank “what do you mean by difficult”. It could mean that the bookkeeping apps they’ve used in the past are not user-friendly, but it could also mean that inputting their transactions is time-consuming, or it could be something entirely different. If you focus on your pre-supposed assumptions of the interviewee’s experience, you will miss out on key insights.
Always encourage your interviewee to be specific so you have a clearer picture of their experiences.
5. Ask what they are currently doing to address the problem you’re trying to solve for them
In many situations, a good way to gauge if the problem you’re trying to solve is a painful one is by asking your interviewees what they are currently doing or have tried to do in the past to solve it.
When I was conducting my interviews for the bookkeeping app, several people told me that they would use an app that helps them record their transactions. I always followed up by asking whether they had tried looking up a bookkeeping app in the past. In most situations, they had not. I thus realized that having an app to maintain their balance sheet was likely not an acute need for them.
Asking people if and how they’ve tried to solve the problem you’re tackling is an effective way to gauge how painful said problem actually is for them.
3 things to avoid
1. Avoid asking super-specific questions about features
Doing so has the risk of biasing the interviewee’s response. For example, asking a question like — “would you like your digital calendar to send you reminders before scheduled events?” — will likely result in a “yes”. If you want to understand how important that feature is to the interviewee, you can suggest a few features and ask them if they had to choose just 1, which one it would be. Another approach is to ask them to rank the options.
2. Avoid suggesting answers to open-ended questions
We are often uncomfortable with long silences and sometimes feel a need to help the interviewee by suggesting a few responses. In these situations, we are actually rushing them to give an answer, bias them, and likely miss out on valuable insights. It’s ok to let your interviewees think! Smile and keep an open stance to make them comfortable. It’s really ok.
3. Avoid interviewing multiple people at the same time
If you interview multiple people at the same time, you run the risk of having them bias each other. For example, earlier this year I was trying to understand how much college students expected to pay for a rental car daily. My approach was to walk around Stanford’s campus and ask each person I saw that one question: “how much would you pay per day if you were to rent a car?”. Initially, I got excited when I saw a group of students sitting together because I thought that would help me save time by asking them the question all at once. What I realized is that as soon as someone gave what felt like a reasonable answer (not necessarily the first answer), everyone else followed it.
When conducting your customer discovery interviews, try your best to avoid biasing the responses of your interviewees, and focus more on past behaviors instead of general feelings. If you’re exploring or working on an idea, I encourage you to conduct as many of these interviews as you can and do so continuously. The goal is to be constantly learning your potential or current customers’ needs and behaviors. Don’t get outlearned by someone else who could potentially disrupt your idea.
