Reading:
How to run user interviews as a startup founder.

Image

How to run user interviews as a startup founder.

January 21, 2020

Here are a few lessons I have learned about doing user interviews as a founder.

#1 The definition of users will keep changing

Our definition of who our users are (or going to be) keeps changing. Starting from something like “all restaurant owners” to “restaurant owners in Singapore” to “restaurant owners in Singapore who run bakeries that sell designer cupcakes, and have one outlet”.

That’s cool because narrowing down into specific segments is a good thing. It means we can reduce the scope of what it will take to make them happy. That’s the way to make them happy faster.

Once we figure out how to satisfy “the outlets that sell designer cupcakes”, we can figure out how to do the same for the outlets that specialize in cakes, then pastries, then bakeries, etc.

It’s impossible to “figure things out” and “get shit done” at the same time so the most optimal way is to separate the two — figure something small out, and get it done. There’s no need to worry about “making all restaurant owners” happy if you can’t make one happy.

#2 Having a really narrow definition is okay

After accepting that the definition of the user is going to keep changing, it’s best to start by talking to the ones who fit in your current definition. If you’ve already built something, show it to everyone who fits your definition, and find the one, two or five people who “get” it.

Dig deep into why they “get” it — What’s the problem they’re trying to solve with it? What do they hope to achieve with it? What improvements will they have in their life because of it? How will they measure it? Is there anyone else who they think will get it too? If so, who? Can they connect you to them? Talk to more of them.

#3 Listen to users, seriously

Show them whatever you’re working on, whether it’s on a napkin, a presentation, a prototype or a product demo, doesn’t matter. And then start listening. Two things can happen, either they start talking about “this is how I could see using it”, or “this is how I could see other people using it”. If it’s the first, dig deeper.

Most times it’ll be the second one, the right question at the time is “why can’t you see yourself using it?”. And they’ll tell you exactly why not — those reasons are what they’re looking for in a solution, their problems, their needs. Note them down, they’re going to come in really handy later.

#4 Ask your users who you should talk to next.

At the end of each conversation, ask the person you speak with — who else you should talk to about your product or idea. You’ve spent time explaining what you do/have in mind, and they know people.

Most times they’d be happy to connect you — after all they did say they could see other people using what you’re talking about. And it’s the quickest way to get to the right people.

#5 Ask questions that will hurt your feelings

What do you really think about what I’m showing you? If they say they love it, ask them to give you money upfront, or write a testimonial, or sign a letter of intent, or get on a call with investors to tell them your idea or product is awesome. If they don’t, ask them why not? And they’ll tell you their problems with it.

The thing to remember is that the only way your startup is going to work out is if you’re able to really help someone out. And there are only two things people really care about “time” or “money”. Either you can save them time or make them money.

#6 Figure out how to save them time, or make them money

The world is running fine without our startups existing. If our goal is to change it in some way and make it better, we must figure out how it is currently getting run. The baker in this story is selling his designer cupcakes and running his business.

We can either help him make his cupcakes cheaper, or sell them to more people. To be able to do either of those, we need to understand each step of his business, evaluate them for inefficiencies and improve them in some way.

It could be the way he’s taking his orders or the way he’s reaching out to people walking by. As long as there is room for improvement, and the cost of switching (the effort they need to put in to use your solution) is lesser, it’s likely they’d give it a go.

The formal name for this way of thinking is “Jobs to be done” — definitely worth a google search.

#7 Listen until you can accurately predict what you’ll hear

When you’ve found a bunch of people, who just seem to “get it” — and every new conversation you’re having seems to be repetitive. You keep hearing the same thing again and again. You’ve found the set of people to go after. Define what’s common between them, and figure out how to address it.

#8 If it feels like you’re going in circles, you’re probably not listening.

People can say the same thing in a million different ways. “It’s really hard to train new bakers”, “getting the colors mixed right takes a lot of practice”, “the first time I ever tried this, it was a disaster”, etc.

It’s really easy to miss these subtle, but critical cues in conversation. But if you listen, and think about them a little more, you may find they’re all talking about a specific task in their process that’s hard to get right. Or a problem they care about.

#9 Listening actively takes practice.

Getting good at picking these up takes a lot of practice, but in general, two simple things can help quite a bit. First taking super detailed notes, or if you’re going alone, recording those conversations.

And second, taking the time to reflect on those notes, so you can look out for things you might have missed out on. Believe me, they’re there, we just missed them because we were too busy thinking about how awesome our idea is going to be for them.

#10 Emotions indicate something worth digging into.

Pay close attention when you sense that the person is emotionally engaged in the conversation. It’s an indicator that they care about something. And people care about things that affect time and money.

You’re probably going to find some gold in there.

The key is to stop thinking about what you’ve got in your head and just ask questions that keep them talking. It’s very easy to distract people from a chain of thought, so when they’re saying stuff they care about, it’s best just to nod and go “hmmm…” to keep them going.

#11 If you’re not 100% confident about your next steps, you haven’t talked to enough users.

Each time you’re guessing why something isn’t working, or what you should be doing next. The default answer is talking to users. Even though it might feel counterproductive, time spent talking to users never goes to waste.

It’s way better to do that than to spend time guessing what they want. Building stuff takes time, experimentation without research is just trial and error. And the odds of finding something that works go up significantly after talking to users.

#12 Everyone is nice until you ask them for something in return

To know if you’ve really got something, always ask the user for something in return. Whether it’s their email address, a referral, a cheque, or a recommendation — no one, other than your friends and family will give you something if it doesn’t bring them some value in return.

The more valuable it is for them, the more they’ll be willing to give you in return.


If you’re currently doing, or about to start doing some user interviews, you can join our beta for free. It’ll help you make sense of notes from your user interviews.

And if you want to keep reading here’s another link to learn more about how to analyse user interviews.


Epiphany is a collaborative discovery platform for product teams. It helps analyze qualitative data from user research, to find customer insights that can drive the product roadmap.

You can use it to take notes, categorize your observations, share your insights with the world and validate them with data.

About me: I’m Arnav you can reach me over here.

Join our beta

Find and share insights from qualitative data faster.

Request Access

Related Stories

Insights in qualitative data
February 4, 2020

From qualitative data to insights

By gathering qualitative data from users, product designers can make out what is available in the market, what the users want and how to fill the gap.

What are insights?
November 18, 2019

What are insights?

Arrow-up