Have you built something based on positive customer feedback, only to see it flop? Have you been unsure about who to listen to? Or even where to start?
I’ve been there many times. There are two types of flawed beliefs:
- Customers don’t know what they want. There’s little use in talking to them
- Customers know exactly what they want. You should build whatever they ask
The reality is that customers know their own problems better than anyone else. But to uncover those problems, you need to ask the right questions. Only then can you imagine and invent solutions on their behalf.
Even though I’ve done hundreds of interviews, I still fall for some common traps.
Trap 1: Ask customers for their opinions on your idea
Reality: This will almost always get you polite, warm, and rather useless feedback. Especially if they know you / like you / you seem eager for approval.
Trap 2: Tell them how your solution works
Reality: Showing always beats telling. By observing what clicks vs. doesn’t, you refine your own intuition.
Trap 3: You need a majority of customers to agree
Reality: You want to see a number of similar customers agree on the problem. When you’re getting lots of contradictory opinions, you may be targeting too broad a customer segment, and / or focusing on abstract solutions.
Our desires to feel validated and seek consensus fog up the truth. Here are some ways to regain clarity.
Know why you’re doing it
There are typically three reasons to do a customer interview:
- Open-ended exploration
- Validate a problem
- Validate a solution
All three are meant to reduce your market risk (aka the dread of building something nobody wants). At each stage, you need questions that tackle your greatest risks.
Effective questions serve as sharp spades for unearthing the truth.
Make it about them, not your idea
When you’re exploring or validating a problem, it’s tempting to pitch your idea. Examples:
- “Sounds like you struggle with quickly reordering products, right?”
- “Can I get your honest thoughts on an idea I’m very excited about?”
- “I’ve been working on a new express reorder flow. Can I tell you more about it?”
These words prime people to be more agreeable. Very few want to rain on your parade. Rather than collect false positives, use early interviews as an opportunity to learn about your customers. Examples:
- “Can you walk me through how you run your business?”
- “Sounds like reordering is a common action. Why do you do it?”
- “Tell me about the last time you reordered. How was the experience?”
- “You mentioned it’s hard to keep track of reorders. How does it impact your business?”
The best way to learn new, concrete facts is to ask for specific examples, and figure out why customers think and feel a certain way. You want to reconstruct their worldview so you can imagine and invent on their behalf. Each interview should make that worldview fuller and crisper. Questions help you continuously fill in the blanks.
Making it about them also creates a more lively conversation. People love talking about the minutiae of their lives. It helps if you’re not a complete stranger, which is where warm leads come in handy, but if you really need a hook, something like this is surprisingly effective: “I’m interested in X, and would love to learn from you so I can build something useful.”
Beware of the focusing illusion
It’s best to start with very open-ended questions before focusing on anything specific. Zoom in too early, and you’ll find yourself chasing a mirage.
In the words of Daniel Kahneman, “Nothing in life is as important as you think it is while you are thinking about it.” When you spend all of your time on a specific problem, you fool yourself and the customer into thinking it must be a big deal.
The world is filled with problems that do not matter. Questions to escape this trap:
- “If we take a step back, how does reordering compare to all the other problems you face?”
- “How do you currently solve the problem of tracking reorders?”
- “Have you looked for better solutions? How much have you spent on them?”
The best evidence of a real problem is that people have spent time / money to fix it, perhaps have created their own workarounds, and are still unhappy. Check that it’s not merely something fun to complain about with no urgency to fix.
What if you’re solving a problem nobody knew they had until you showed what they’re missing? Just in time for the next step.
Show, don’t tell
Once you’ve passed the problem validation stage, you do want to show your idea. The fastest way for someone to “get” your vision is to play with your designs / prototypes. Prototypes > static images > words. This stage is especially important if turning the prototype into a real product will take a lot of effort.
If the customer is struggling, resist the urge to immediately jump in and explain yourself. Ask them to put their thoughts on speakerphone, so you get unfiltered feedback to iterate on. The best evidence of a useful solution is when customers intuitively grasp how it works, then ask to start paying / using / sharing it.
Actions speak louder than words. And there is no louder action than getting out your wallet and sharing the product with other people.
It will take a lot of work to get to this aha moment. You may even need more of a marketing / sales push to lock them in. That’s ok, but remember your goal is to ultimately craft an offering that is seen as a no-brainer.
This stage makes or breaks you. Execution is the only vision and strategy your customer will ever experience.
Here are some other questions best avoided:
- “What would you pay for this?” A better anchor is to understand the value your product provides. How do they currently solve the problem? How much does it cost in time and money? Can they get a 10X ROI from your product?
- “Could you ever see yourself using this?” Don’t bet on the remote possible :)
- “On a scale of 1-10, how likely are you to recommend this?” Better to save this for a survey where you can get more data points. Interviews allow to collect deep context from a handful of people. Use it wisely!
If you’re working on a team, it helps you to have a process for sharing customer learnings so you don’t become the bottleneck. Context not control is the best way to scale high standards.
Some simple steps to get there:
- Ask the team to submit questions for the interview guide
- Have at least 2 people at the interview. One person asks questions, the other takes notes and can jump in if something is missed. If the customer is ok with it, record the call
- Share notes and recordings; notes are ideally verbatim, with takeaways of new learnings
- Meet as a team to update collective beliefs (as needed)
Unless you’re building something purely for yourself, you will benefit from talking to customers. The magic behind “product intuition” is simply understanding customers so well you can anticipate how they would react.
To reach that clarity, get to know them before you sell splashy ideas, and use every interview as an opportunity to update your most valuable learnings.
- The Mom Test: How to talk to customers & learn if your business is a good idea when everyone is lying to you — must-read, especially if you’re working on 0 to 1 products. Packed with funny examples of dialogues gone right and wrong
- A new approach to feature requests - detailed walkthrough of how to ask questions to figure out why a customer wanted a feature
- Why "validate your idea" is bad advice