Customer discovery interviews are one of the highest impact tactics a founder can use to ensure the product they plan to build will actually resonate with its target market. But just “getting out of the building and talking to people,” as the startup adage goes, isn’t going to get you the results you need. If you do these interviews without a plan, you inevitably end up doing most of the talking and don’t walk away with any useful data.
The most experienced founders know that a customer research interview guide is the key to not only keeping your customer discovery sessions on track but to gather insights that will ultimately benefit your future product development. Rob Hayes, who’s the founder of Foundation.pm, says that “a good guide is arguably the most important part of your testing plan.” And Jordan Clayton, who’s done the research and product design for companies like Jibo, TripAdvisor, and Ubisoft, compares conducting research without a guide to “going into a presentation without preparing.”
But how do you actually create a research interview guide, what do you put in it, and how do you use it when your interview gets off-track?
Here’s what I’ve learned from creating my own discussion guide for Grain, plus what several experts told me about their own experience testing and refining their research interview guides across hundreds of sessions.
First, we have to understand the purpose of customer discovery (hint: it’s not to make you feel good). Founders who have never done customer discovery research sometimes misunderstand the purpose of these interviews. So let’s clear that up right now.
The discovery phase does not exist to confirm your suspicions about a product idea — to help you say “yes, the market really does need this because a bunch of people told me they liked my idea.” You should not expect to feel validated and praised — actually, expect the opposite. Truth be told, if you walk away from these interviews feeling nothing but great about yourself, there’s a 99% chance you did something wrong.
The purpose of customer discovery interviews is to poke holes in your idea so that you can continuously refine it. You’re basically asking your interviewees to expose any flaws or imperfections in your thinking. You want them to tear those flaws wide open to expose the potential challenges or barriers with your idea before you start developing.
Before writing an interview guide, you must accept that customer discovery doesn’t exist to validate your idea. Instead, it is meant to refine your idea and shape it into something that stands a chance of resonating with your ideal customer base. Now you can decide which questions to ask.
Next, write unbiased, reality-based questions. To pick apart the flaws in your idea and refine it into something the market could be excited about, you need to ask some really good questions. And you need to do so carefully to avoid your own biases.
Challenging your preconceived notions and theories means you cannot “lead the witness.” In other words, you can’t ask questions in such a way that the interviewee feels compelled to answer one way more than another. So, let’s say your product helps event organizers find available spaces for rent. Instead of asking an interviewee to tell you a story about a time they couldn’t locate a meeting space to hold a workshop, you’d ask her to tell you about a recent challenge they had when organizing a workshop. If finding a meeting room was indeed a problem, she’ll tell you.
This bias can turn into a costly mistake if you are the only one doing the customers research interviews. As Rob Fitzpatrick, author of The Mom Test calls, you’ll end up becoming “a single point of failure”
Avoid leading the witness and other common biases by writing questions for your research interview guide that are both open-ended and grounded in reality.
Your interview questions should not be easily answered with “yes” or “no.” They should be thought-provoking and encourage the interviewee to talk about their behavior at length, so you get the context and depth of information required to make an accurate decision about your own product idea.
For example, let’s say you want to build an alternative to Slack, but don’t understand how often your target market actually uses it. An open-ended question appropriate for your research might be “how often do you use Slack?” It’s not "do you use Slack," which can easily be answered with a "yes" and a blank stare — it’s an open-ended question that gets your interviewee thinking and talking.
Of course, the scenario above assumes your target audience is already using Slack. To avoid leading the witness otherwise, you could ask something more like, “Which tools do you use to communicate asynchronously at work?”
Try adding open-ended questions similar to these to your customer discovery discussion guide:
One final tip about these open-ended questions: if a user is giving short answers, or you’re not sure you fully understood what they shared, repeat what they’ve said back to them. As founder and executive coach Whitney Hess says:
Things get lost in translation even when you’re both speaking the same language. So, as you’re going along in your interview, take the time to summarize a key learning, and repeat it back to the participant. This gives them the chance to confirm or clarify and will keep you from going down the wrong path later on.
You may also find that making the interviewee repeat themselves will cause them to elaborate a little more, surfacing details you would have otherwise missed.
Your questions must refer to real-life incidents — what users are currently doing, what they’ve done in the past, and how they feel about these experiences.
No hypotheticals. No imagining or speculating. Don’t ask your interviewee to theorize. And avoid questions that start with “What would you do if...”
Because speculative questions aren’t grounded in reality, even if a user says they’re absolutely sure they would (hypothetically) use a product, they’re just guessing. Focus only on what they have already experienced and felt. Like this User Interviews guide says, “Past behavior is the best indication of future behavior.”
Jason Quey, founder of Growth Ramp, agrees:
Because humans cannot predict the future, you can only be certain of what has happened to someone in the past. And by asking open-ended questions, the customer can tell you in their own words what matters most to them.
Proven behavior — what someone is already doing — is always going to be the best possible indicator of how your target users will react to a situation.
Remember our Slack question above? You can take it one step further by grounding it in your customer’s reality: “Tell me about a time when you used Slack to solve a problem at work.”
There’s a ton you can draw out of this story: how this person (who’s presumably in your target audience) used Slack, how they approached problem-solving, and maybe even how often they use Slack specifically for solutions to complex work issues.
Here are a few specific questions that’ll get your interviewees talking about their real experiences, problems, and needs:
Use your guide to get to the root of your customers’ problems — then keep digging. Even with the best-laid plans, your customer discovery interviews can get off track. This is OK. In fact, the folks I talk to say it’s one of the best things that can happen during these interviews.
As Michael Probert, Head of Growth at Uptick told me, “Some of the deepest learnings can only be found by interviewing, catching a glimpse of frustration, and then pulling on that thread to see what's at the root.”
Scott McCall, who's currently the UX Design Lead at IBM, agrees:
There’s a lot of stuff that can’t be put into a research guide. Great interviewers will pull things out [of answers], and use things like silence as a tool. All these things can’t really be readily put into a guide. Above all else, you’re trying to make the person feel comfortable and allow them to really be thoughtful about the responses.
Your research interview guide should give you the structure you need to ask a series of solid questions. But you, as the interviewer, should be prepared to go off-guide — to keep digging until you get to the root of a customer’s real problems or needs.
If an interviewee reacts very strongly to a question — with frustration, sadness, joy, confusion, and so on — you’re going to want them to tell you more. You want to understand why they feel that way. Like I said earlier, understanding someone’s current behavior and feelings is the key to understanding how your product idea might help them. But you may have to “pull on the thread,” as Michael Probert says, to get an interviewee to keep talking.
Here are what the experts I spoke to suggested you do when you want to dig further into your customers’ problems and expose some truly valuable feedback.
You may be surprised to learn how effective silence can be as a thread-puller. Malini Jagannadhan, who leads Product at Curalate, suggests you get accustomed to uncomfortable silence as a part of this “digging.” Sometimes interviewees can take a while to formulate their thoughts or react in a way that’s more emotional. Malini knows it’s a struggle to stay quiet and watch an interviewee work out what they’re really thinking, but she says, “You have to get comfortable with leaving that gap and letting people work through it.”
If the silence drags out too long, Malini suggests open-ended follow-up questions — like, “What are you thinking right now?” — that encourage the interviewee, but that doesn’t lead them toward any one direction. She emphasized that you want the people you're speaking with to arrive at conclusions naturally, even if that means waiting a while. These interviews are the only way to bring out “unexpected nooks and crannies” in data about your ideal customer, so a little awkward silence is worth it.
On the other hand, interviewees sometimes talk too much, going on tangents about subjects not relevant to your research. If this happens, redirect an interviewee’s attention to a topic you visited earlier — “Tell me more about how you use Airtable.” — or transition to the next question in your guide. Course-correcting will become more natural with every interview you do.
But don’t cut them off too soon. Adam Sigel, who’s the Head of Product at Hometap, suggests that you always listen for emotion in these tangents, just in case they do tie back to your product research. These emotions might be something you want to tap into for follow-up questions that will uncover more important insights:
If a research participant starts talking about their problems with property taxes, or their ideas for their new kitchen renovation, or the struggles of landscaping a backyard, those are all things Hometap [a home equity startup] needs to know about. On the other hand, if they want to spend the entire conversation talking about a fight they had with their mother-in-law, that’s not quite so relevant to Adam’s research and he’ll try to steer them back towards home-related topics.
Finally, recognize that a great research interview guide is always evolving. One final piece of advice: think of your research interview guide as a first draft. Your guide should always be changing to help you have better conversations, even if that means all your initial questions get swapped out or your initial vision of your product changes entirely.
Curalate’s Malini Jagannadhan says they “keep interviewing until [they] find themes” in their data. Do the same with your interview questions, using your video or audio recordings of interviews to look for patterns.
If you find yourself going off-script to ask the same few questions repeatedly, revise your guide to add them in. And most importantly, as your vision of your product changes, remember to change your questions. It’s pointless to collect data on something that’s no longer relevant to your idea. Experiment with new questions instead, take notes and record the sessions to see what works, and put the winners into your guide for long-term usage. All set? Let's put this guide to action — and learn how to conduct user research interviews to build the right product.
Try out our ready-made and tested Zapier templates using Grain. Click the “Use this Zap” to get started.